Skip to contentSkip to conversation

Allocate memmap from hotadded memory (per device)

PATCH In review Oscar Salvador 2 participants 24 messages Dec 17, 2020 – Jan 25, 2021

Subsystems (12)
lore
Share & tools
lore

Oscar Salvador

author maintainer
#0 lore raw

cover Allocate memmap from hotadded memory (per device)

I figured I would send a new version before going on vacation, so I can work on it when I am back.

Changes from RFCv3 to Patchv1:
- Addressed feedback from David
- Re-order patches

Changes from v2 -> v3:
- Re-order patches (Michal)
- Fold "mm,memory_hotplug: Introduce MHP_MEMMAP_ON_MEMORY" in patch#1
- Add kernel boot option to enable this feature (Michal)

Changes from v1 -> v2:
- Addressed feedback provided by David
- Add a arch_support_memmap_on_memory to be called from mhp_supports_memmap_on_memory, as atm, only ARM, powerpc and x86_64 have altmat support.

Original cover letter:

----

The primary goal of this patchset is to reduce memory overhead of the hot-added memory (at least for SPARSEMEM_VMEMMAP memory model).
The current way we use to populate memmap (struct page array) has two main drawbacks:

a) it consumes an additional memory until the hotadded memory itself is onlined and
b) memmap might end up on a different numa node which is especially true for movable_node configuration.
c) due to fragmentation we might end up populating memmap with base pages

One way to mitigate all these issues is to simply allocate memmap array (which is the largest memory footprint of the physical memory hotplug) from the hot-added memory itself. SPARSEMEM_VMEMMAP memory model allows us to map any pfn range so the memory doesn't need to be online to be usable for the array. See patch 3 for more details.
This feature is only usable when CONFIG_SPARSEMEM_VMEMMAP is set.

[Overall design]:

Implementation wise we reuse vmem_altmap infrastructure to override the default allocator used by vmemap_populate. memory_block structure gained a new field called nr_vmemmap_pages. This plays well for two reasons:

1) {offline/online}_pages know the difference between start_pfn and buddy_start_pfn, which is start_pfn + nr_vmemmap_pages. In this way all isolation/migration operations are done to within the right range of memory without vmemmap pages. This allows us for a much cleaner handling.

2) In try_remove_memory, we construct a new vmemap_altmap struct with the right information based on memory_block->nr_vmemap_pages, so we end up calling vmem_altmap_free instead of free_pagetable when removing the memory.

Oscar Salvador (5):
mm: Introduce ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE mm,memory_hotplug: Allocate memmap from the added memory range acpi,memhotplug: Enable MHP_MEMMAP_ON_MEMORY when supported powerpc/memhotplug: Enable MHP_MEMMAP_ON_MEMORY when supported mm,memory_hotplug: Add kernel boot option to enable memmap_on_memory

 .../admin-guide/kernel-parameters.txt         |  14 ++
 arch/arm64/Kconfig                            |   4 +
 arch/powerpc/Kconfig                          |   4 +
 .../platforms/pseries/hotplug-memory.c        |   5 +-
 arch/x86/Kconfig                              |   4 +
 drivers/acpi/acpi_memhotplug.c                |   5 +-
 drivers/base/memory.c                         |  20 ++-
 include/linux/memory.h                        |   8 +-
 include/linux/memory_hotplug.h                |  21 ++-
 include/linux/memremap.h                      |   2 +-
 include/linux/mmzone.h                        |   5 +
 mm/Kconfig                                    |   3 +
 mm/Makefile                                   |   5 +-
 mm/memory_hotplug.c                           | 158 +++++++++++++++---
 mm/page_alloc.c                               |   4 +-
 15 files changed, 224 insertions(+), 38 deletions(-)
Signature · 1 line
2.26.2
author maintainer
#1 lore raw

PATCH 1/5 mm: Introduce ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE

In order to use self-hosted memmap array, the platform needs to have support for CONFIG_SPARSEMEM_VMEMMAP and altmap.
Currently, only arm64, PPC and x86_64 have the support, so enable those platforms with ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE.

ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE will be checked later on to see whether we can enable this feature or not.

---
 arch/arm64/Kconfig   | 4 ++++
 arch/powerpc/Kconfig | 4 ++++
 arch/x86/Kconfig     | 4 ++++
 mm/Kconfig           | 3 +++
 4 files changed, 15 insertions(+)
arch/arm64/Kconfig +4 -0
index 33446269f692..2daa29ab9057 100644
@@ -307,6 +307,10 @@ config ARCH_ENABLE_MEMORY_HOTPLUG
 config ARCH_ENABLE_MEMORY_HOTREMOVE
 	def_bool y
 
+config ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE
+	def_bool y
+	depends on MEMORY_HOTPLUG && SPARSEMEM_VMEMMAP_ENABLE
+
 config SMP
 	def_bool y
 
arch/powerpc/Kconfig +4 -0
index 107bb4319e0e..30322105f145 100644
@@ -527,6 +527,10 @@ config ARCH_ENABLE_MEMORY_HOTPLUG
 config ARCH_ENABLE_MEMORY_HOTREMOVE
 	def_bool y
 
+config ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE
+        def_bool y
+        depends on MEMORY_HOTPLUG && SPARSEMEM_VMEMMAP_ENABLE
+
 config PPC64_SUPPORTS_MEMORY_FAILURE
 	bool "Add support for memory hwpoison"
 	depends on PPC_BOOK3S_64
arch/x86/Kconfig +4 -0
index 42d84d86f1f4..742b29169b47 100644
@@ -2443,6 +2443,10 @@ config ARCH_ENABLE_MEMORY_HOTREMOVE
 	def_bool y
 	depends on MEMORY_HOTPLUG
 
+config ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE
+	def_bool y
+	depends on X86_64 && MEMORY_HOTPLUG && SPARSEMEM_VMEMMAP_ENABLE
+
 config USE_PERCPU_NUMA_NODE_ID
 	def_bool y
 	depends on NUMA
mm/Kconfig +3 -0
index 7af1a55b708e..6156114b7974 100644
@@ -183,6 +183,9 @@ config MEMORY_HOTREMOVE
 	depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE
 	depends on MIGRATION
 
+config ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE
+	bool
+
 # Heavily threaded applications may benefit from splitting the mm-wide
 # page_table_lock, so that faults on different parts of the user address
 # space can be handled with less contention: split it at this NR_CPUS.
Signature · 1 line
2.26.2
#2 lore raw

Re: [PATCH 1/5] mm: Introduce ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE

Quoted 4 lines · Oscar Salvador

In order to use self-hosted memmap array, the platform needs to have support for CONFIG_SPARSEMEM_VMEMMAP and altmap.
Currently, only arm64, PPC and x86_64 have the support, so enable those platforms with ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE.

"In the first version, only .... will support it".

I'd probably split off enabling it per architecture in separate patches and the end of the series.

Apart from that looks good.

Quoted 5 lines

ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE will be checked later on to see whether we can enable this feature or not.

Signature · 3 lines
Thanks,

David / dhildenb
author maintainer
#3 lore raw

Re: [PATCH 1/5] mm: Introduce ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE

Quoted 7 lines · David Hildenbrand
Quoted 4 lines · Oscar Salvador

In order to use self-hosted memmap array, the platform needs to have support for CONFIG_SPARSEMEM_VMEMMAP and altmap.
Currently, only arm64, PPC and x86_64 have the support, so enable those platforms with ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE.

"In the first version, only .... will support it".

I will try to be more specific.

Quoted 3 lines

I'd probably split off enabling it per architecture in separate patches and the end of the series.

You mean introducing only mm/Kconfig change in this patch, and then arch/*/Kconfig changes in separate patches at the end of the series?

I can certainly do that, not sure how much will help with the review, but it might help when bisecting.

Quoted 1 line

Apart from that looks good.

Thanks for taking a look David!

Signature · 2 lines
Oscar Salvador
SUSE L3
#4 lore raw

Re: [PATCH 1/5] mm: Introduce ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE

Quoted 9 lines · Oscar Salvador
Quoted 7 lines · David Hildenbrand
Quoted 4 lines · Oscar Salvador

In order to use self-hosted memmap array, the platform needs to have support for CONFIG_SPARSEMEM_VMEMMAP and altmap.
Currently, only arm64, PPC and x86_64 have the support, so enable those platforms with ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE.

"In the first version, only .... will support it".

Quoted 8 lines

I will try to be more specific.

Quoted 3 lines

I'd probably split off enabling it per architecture in separate patches and the end of the series.

You mean introducing only mm/Kconfig change in this patch, and then arch/*/Kconfig changes in separate patches at the end of the series?

Yeah, or squashing the leftovers of this patch (3 LOC) into patch #2.

Quoted 3 lines

I can certainly do that, not sure how much will help with the review, but it might help when bisecting.

It's usually nicer to explicitly enable stuff per architecture, stating why it works on that architecture (and in the best case, even was tested!). :)

Signature · 3 lines
Thanks,

David / dhildenb
author maintainer
#5 lore raw

Re: [PATCH 1/5] mm: Introduce ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE

Quoted 5 lines · David Hildenbrand
Quoted 2 lines · Oscar Salvador

You mean introducing only mm/Kconfig change in this patch, and then arch/*/Kconfig changes in separate patches at the end of the series?

Yeah, or squashing the leftovers of this patch (3 LOC) into patch #2.

Ok, makes sense.

Quoted 6 lines
Quoted 2 lines

I can certainly do that, not sure how much will help with the review, but it might help when bisecting.

It's usually nicer to explicitly enable stuff per architecture, stating why it works on that architecture (and in the best case, even was tested!). :)

Fine by me.
I will prepare another re-spin with that in mind then.

It would be great to have some feedback on patch#2 before that (in case you find some time ;-).

Thanks!

Signature · 2 lines
Oscar Salvador
SUSE L3
#6 lore raw

Re: [PATCH 1/5] mm: Introduce ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE

Quoted 16 lines · Oscar Salvador
Quoted 5 lines · David Hildenbrand
Quoted 2 lines · Oscar Salvador

You mean introducing only mm/Kconfig change in this patch, and then arch/*/Kconfig changes in separate patches at the end of the series?

Yeah, or squashing the leftovers of this patch (3 LOC) into patch #2.

Ok, makes sense.

Quoted 6 lines
Quoted 2 lines

I can certainly do that, not sure how much will help with the review, but it might help when bisecting.

It's usually nicer to explicitly enable stuff per architecture, stating why it works on that architecture (and in the best case, even was tested!). :)

Quoted 5 lines

Fine by me.
I will prepare another re-spin with that in mind then.

It would be great to have some feedback on patch#2 before that (in case you find some time ;-).

Yeah, will get to that soon. Mailbox is flooded right now :D

Signature · 3 lines
Thanks,

David / dhildenb
author maintainer
#7 lore raw

PATCH 2/5 mm,memory_hotplug: Allocate memmap from the added memory range

In reply to Oscar Salvador

Physical memory hotadd has to allocate a memmap (struct page array) for the newly added memory section. Currently, alloc_pages_node() is used for those allocations.

This has some disadvantages:
a) an existing memory is consumed for that purpose

    (eg: ~2MB per 128MB memory section on x86_64)

b) if the whole node is movable then we have off-node struct pages which has performance drawbacks.
c) It might be there are no PMD_ALIGNED chunks so memmap array gets populated with base pages.

This can be improved when CONFIG_SPARSEMEM_VMEMMAP is enabled.

Vmemap page tables can map arbitrary memory.
That means that we can simply use the beginning of each memory section and map struct pages there.
struct pages which back the allocated space then just need to be treated carefully.

Implementation wise we will reuse vmem_altmap infrastructure to override the default allocator used by __populate_section_memmap.
Part of the implementation also relies on memory_block structure gaining a new field which specifies the number of vmemmap_pages at the beginning. This comes in handy as in {online,offline}_pages, all the isolation and migration is being done on (buddy_start_pfn, end_pfn] range, being buddy_start_pfn = start_pfn + nr_vmemmap_pages.

In this way, we have:

(start_pfn, buddy_start_pfn - 1] = Initialized and PageReserved
(buddy_start_pfn, end_pfn] = Initialized and sent to buddy

Hot-remove:

We need to be careful when removing memory, as adding and removing memory needs to be done with the same granularity. To check that this assumption is not violated, we check the memory range we want to remove and if a) any memory block has vmemmap pages and b) the range spans more than a single memory block, we scream out loud and refuse to proceed.

If all is good and the range was using memmap on memory (aka vmemmap pages), we construct an altmap structure so free_hugepage_table does the right thing and calls vmem_altmap_free instead of free_pagetable.

---
 drivers/base/memory.c          |  20 +++--
 include/linux/memory.h         |   8 +-
 include/linux/memory_hotplug.h |  21 ++++-
 include/linux/memremap.h       |   2 +-
 include/linux/mmzone.h         |   5 ++
 mm/memory_hotplug.c            | 138 ++++++++++++++++++++++++++++-----
 mm/page_alloc.c                |   4 +-
 7 files changed, 163 insertions(+), 35 deletions(-)
drivers/base/memory.c +12 -8
index eef4ffb6122c..7219e805dcaa 100644
@@ -175,7 +175,7 @@ int memory_notify(unsigned long val, void *v)
  */
 static int
 memory_block_action(unsigned long start_section_nr, unsigned long action,
-		    int online_type, int nid)
+		    int online_type, int nid, unsigned long nr_vmemmap_pages)
 {
 	unsigned long start_pfn;
 	unsigned long nr_pages = PAGES_PER_SECTION * sections_per_block;
@@ -185,10 +185,11 @@ memory_block_action(unsigned long start_section_nr, unsigned long action,
 
 	switch (action) {
 	case MEM_ONLINE:
-		ret = online_pages(start_pfn, nr_pages, online_type, nid);
+		ret = online_pages(start_pfn, nr_pages, nr_vmemmap_pages,
+				   online_type, nid);
 		break;
 	case MEM_OFFLINE:
-		ret = offline_pages(start_pfn, nr_pages);
+		ret = offline_pages(start_pfn, nr_pages, nr_vmemmap_pages);
 		break;
 	default:
 		WARN(1, KERN_WARNING "%s(%ld, %ld) unknown action: "
@@ -211,7 +212,7 @@ static int memory_block_change_state(struct memory_block *mem,
 		mem->state = MEM_GOING_OFFLINE;
 
 	ret = memory_block_action(mem->start_section_nr, to_state,
-				  mem->online_type, mem->nid);
+				  mem->online_type, mem->nid, mem->nr_vmemmap_pages);
 
 	mem->state = ret ? from_state_req : to_state;
 
@@ -571,7 +572,8 @@ int register_memory(struct memory_block *memory)
 	return ret;
 }
 
-static int init_memory_block(unsigned long block_id, unsigned long state)
+static int init_memory_block(unsigned long block_id, unsigned long state,
+			     unsigned long nr_vmemmap_pages)
 {
 	struct memory_block *mem;
 	unsigned long start_pfn;
@@ -591,6 +593,7 @@ static int init_memory_block(unsigned long block_id, unsigned long state)
 	start_pfn = section_nr_to_pfn(mem->start_section_nr);
 	mem->phys_device = arch_get_memory_phys_device(start_pfn);
 	mem->nid = NUMA_NO_NODE;
+	mem->nr_vmemmap_pages = nr_vmemmap_pages;
 
 	ret = register_memory(mem);
 
@@ -610,7 +613,7 @@ static int add_memory_block(unsigned long base_section_nr)
 	if (section_count == 0)
 		return 0;
 	return init_memory_block(memory_block_id(base_section_nr),
-				 MEM_ONLINE);
+				 MEM_ONLINE, 0);
 }
 
 static void unregister_memory(struct memory_block *memory)
@@ -632,7 +635,8 @@ static void unregister_memory(struct memory_block *memory)
  *
  * Called under device_hotplug_lock.
  */
-int create_memory_block_devices(unsigned long start, unsigned long size)
+int create_memory_block_devices(unsigned long start, unsigned long size,
+				unsigned long vmemmap_pages)
 {
 	const unsigned long start_block_id = pfn_to_block_id(PFN_DOWN(start));
 	unsigned long end_block_id = pfn_to_block_id(PFN_DOWN(start + size));
@@ -645,7 +649,7 @@ int create_memory_block_devices(unsigned long start, unsigned long size)
 		return -EINVAL;
 
 	for (block_id = start_block_id; block_id != end_block_id; block_id++) {
-		ret = init_memory_block(block_id, MEM_OFFLINE);
+		ret = init_memory_block(block_id, MEM_OFFLINE, vmemmap_pages);
 		if (ret)
 			break;
 	}
include/linux/memory.h +7 -1
index 439a89e758d8..f09d60ef8ad9 100644
@@ -30,6 +30,11 @@ struct memory_block {
 	int phys_device;		/* to which fru does this belong? */
 	struct device dev;
 	int nid;			/* NID for this memory block */
+	unsigned long nr_vmemmap_pages;	/*
+					 * Number of vmemmap pages. These pages
+					 * lay at the beginning of the memory
+					 * block.
+					 */
 };
 
 int arch_get_memory_phys_device(unsigned long start_pfn);
@@ -81,7 +86,8 @@ static inline int memory_notify(unsigned long val, void *v)
 #else
 extern int register_memory_notifier(struct notifier_block *nb);
 extern void unregister_memory_notifier(struct notifier_block *nb);
-int create_memory_block_devices(unsigned long start, unsigned long size);
+int create_memory_block_devices(unsigned long start, unsigned long size,
+				unsigned long vmemmap_pages);
 void remove_memory_block_devices(unsigned long start, unsigned long size);
 extern void memory_dev_init(void);
 extern int memory_notify(unsigned long val, void *v);
include/linux/memory_hotplug.h +17 -4
index 15acce5ab106..67ea4a0b25bd 100644
@@ -70,6 +70,14 @@ typedef int __bitwise mhp_t;
  */
 #define MEMHP_MERGE_RESOURCE	((__force mhp_t)BIT(0))
 
+/*
+ * We want memmap (struct page array) to be self contained.
+ * To do so, we will use the beginning of the hot-added range to build
+ * the page tables for the memmap array that describes the entire range.
+ * Only selected architectures support it with SPARSE_VMEMMAP.
+ */
+#define MHP_MEMMAP_ON_MEMORY   ((__force mhp_t)BIT(1))
+
 /*
  * Extended parameters for memory hotplug:
  * altmap: alternative allocator for memmap array (optional)
@@ -113,11 +121,13 @@ extern int zone_grow_waitqueues(struct zone *zone, unsigned long nr_pages);
 extern int add_one_highpage(struct page *page, int pfn, int bad_ppro);
 /* VM interface that may be used by firmware interface */
 extern int online_pages(unsigned long pfn, unsigned long nr_pages,
-			int online_type, int nid);
+			unsigned long nr_vmemmap_pages, int online_type,
+			int nid);
 extern struct zone *test_pages_in_a_zone(unsigned long start_pfn,
 					 unsigned long end_pfn);
 extern void __offline_isolated_pages(unsigned long start_pfn,
-				     unsigned long end_pfn);
+				     unsigned long end_pfn,
+				     unsigned long buddy_start_pfn);
 
 typedef void (*online_page_callback_t)(struct page *page, unsigned int order);
 
@@ -312,7 +322,8 @@ static inline void pgdat_resize_init(struct pglist_data *pgdat) {}
 #ifdef CONFIG_MEMORY_HOTREMOVE
 
 extern void try_offline_node(int nid);
-extern int offline_pages(unsigned long start_pfn, unsigned long nr_pages);
+extern int offline_pages(unsigned long start_pfn, unsigned long nr_pages,
+			 unsigned long nr_vmemmap_pages);
 extern int remove_memory(int nid, u64 start, u64 size);
 extern void __remove_memory(int nid, u64 start, u64 size);
 extern int offline_and_remove_memory(int nid, u64 start, u64 size);
@@ -320,7 +331,8 @@ extern int offline_and_remove_memory(int nid, u64 start, u64 size);
 #else
 static inline void try_offline_node(int nid) {}
 
-static inline int offline_pages(unsigned long start_pfn, unsigned long nr_pages)
+static inline int offline_pages(unsigned long start_pfn, unsigned long nr_pages,
+				unsigned long nr_vmemmap_pages)
 {
 	return -EINVAL;
 }
@@ -364,6 +376,7 @@ extern struct zone *zone_for_pfn_range(int online_type, int nid, unsigned start_
 extern int arch_create_linear_mapping(int nid, u64 start, u64 size,
 				      struct mhp_params *params);
 void arch_remove_linear_mapping(u64 start, u64 size);
+extern bool mhp_supports_memmap_on_memory(unsigned long size);
 #endif /* CONFIG_MEMORY_HOTPLUG */
 
 #endif /* __LINUX_MEMORY_HOTPLUG_H */
include/linux/memremap.h +1 -1
index 86c6c368ce9b..3de5d482ac1a 100644
@@ -17,7 +17,7 @@ struct device;
  * @alloc: track pages consumed, private to vmemmap_populate()
  */
 struct vmem_altmap {
-	const unsigned long base_pfn;
+	unsigned long base_pfn;
 	const unsigned long end_pfn;
 	const unsigned long reserve;
 	unsigned long free;
include/linux/mmzone.h +5 -0
index b593316bff3d..d16b4411e422 100644
@@ -407,6 +407,11 @@ enum zone_type {
 	 *    techniques might use alloc_contig_range() to hide previously
 	 *    exposed pages from the buddy again (e.g., to implement some sort
 	 *    of memory unplug in virtio-mem).
+	 * 6. Memory-hotplug: when using memmap_on_memory and onlining the memory
+	 *    to the MOVABLE zone, the vmemmap pages are also placed in such
+	 *    zone. Such pages cannot be really moved around as they are
+	 *    self-stored in the range, but they are treated as movable when
+	 *    the range that they describe is about to be offlined.
 	 *
 	 * In general, no unmovable allocations that degrade memory offlining
 	 * should end up in ZONE_MOVABLE. Allocators (like alloc_contig_range())
mm/memory_hotplug.c +118 -20

293 lines not on this page · raw message

mm/page_alloc.c +3 -1
index 774542e1483e..02bca3b880a4 100644
@@ -8825,7 +8825,8 @@ void zone_pcp_reset(struct zone *zone)
  * All pages in the range must be in a single zone, must not contain holes,
  * must span full sections, and must be isolated before calling this function.
  */
-void __offline_isolated_pages(unsigned long start_pfn, unsigned long end_pfn)
+void __offline_isolated_pages(unsigned long start_pfn, unsigned long end_pfn,
+			      unsigned long buddy_start_pfn)
 {
 	unsigned long pfn = start_pfn;
 	struct page *page;
@@ -8836,6 +8837,7 @@ void __offline_isolated_pages(unsigned long start_pfn, unsigned long end_pfn)
 	offline_mem_sections(pfn, end_pfn);
 	zone = page_zone(pfn_to_page(pfn));
 	spin_lock_irqsave(&zone->lock, flags);
+	pfn = buddy_start_pfn;
 	while (pfn < end_pfn) {
 		page = pfn_to_page(pfn);
 		/*
Signature · 1 line
2.26.2
author maintainer
#8 lore raw

Re: [PATCH 2/5] mm,memory_hotplug: Allocate memmap from the added memory range

Quoted 42 lines · Oscar Salvador

Physical memory hotadd has to allocate a memmap (struct page array) for the newly added memory section. Currently, alloc_pages_node() is used for those allocations.

This has some disadvantages:
a) an existing memory is consumed for that purpose

    (eg: ~2MB per 128MB memory section on x86_64)

b) if the whole node is movable then we have off-node struct pages which has performance drawbacks.
c) It might be there are no PMD_ALIGNED chunks so memmap array gets populated with base pages.

This can be improved when CONFIG_SPARSEMEM_VMEMMAP is enabled.

Vmemap page tables can map arbitrary memory.
That means that we can simply use the beginning of each memory section and map struct pages there.
struct pages which back the allocated space then just need to be treated carefully.

Implementation wise we will reuse vmem_altmap infrastructure to override the default allocator used by __populate_section_memmap.
Part of the implementation also relies on memory_block structure gaining a new field which specifies the number of vmemmap_pages at the beginning. This comes in handy as in {online,offline}_pages, all the isolation and migration is being done on (buddy_start_pfn, end_pfn] range, being buddy_start_pfn = start_pfn + nr_vmemmap_pages.

In this way, we have:

(start_pfn, buddy_start_pfn - 1] = Initialized and PageReserved
(buddy_start_pfn, end_pfn] = Initialized and sent to buddy

Hot-remove:

We need to be careful when removing memory, as adding and removing memory needs to be done with the same granularity. To check that this assumption is not violated, we check the memory range we want to remove and if a) any memory block has vmemmap pages and b) the range spans more than a single memory block, we scream out loud and refuse to proceed.

Quoted 5 lines

If all is good and the range was using memmap on memory (aka vmemmap pages), we construct an altmap structure so free_hugepage_table does the right thing and calls vmem_altmap_free instead of free_pagetable.

Let us refloat this one before it sinks deeper :-)

Signature · 2 lines
Oscar Salvador
SUSE L3
#9 lore raw

Re: [PATCH 2/5] mm,memory_hotplug: Allocate memmap from the added memory range

In reply to Oscar Salvador
Quoted 42 lines · Oscar Salvador

Physical memory hotadd has to allocate a memmap (struct page array) for the newly added memory section. Currently, alloc_pages_node() is used for those allocations.

This has some disadvantages:
a) an existing memory is consumed for that purpose

    (eg: ~2MB per 128MB memory section on x86_64)

b) if the whole node is movable then we have off-node struct pages which has performance drawbacks.
c) It might be there are no PMD_ALIGNED chunks so memmap array gets populated with base pages.

This can be improved when CONFIG_SPARSEMEM_VMEMMAP is enabled.

Vmemap page tables can map arbitrary memory.
That means that we can simply use the beginning of each memory section and map struct pages there.
struct pages which back the allocated space then just need to be treated carefully.

Implementation wise we will reuse vmem_altmap infrastructure to override the default allocator used by __populate_section_memmap.
Part of the implementation also relies on memory_block structure gaining a new field which specifies the number of vmemmap_pages at the beginning. This comes in handy as in {online,offline}_pages, all the isolation and migration is being done on (buddy_start_pfn, end_pfn] range, being buddy_start_pfn = start_pfn + nr_vmemmap_pages.

In this way, we have:

(start_pfn, buddy_start_pfn - 1] = Initialized and PageReserved
(buddy_start_pfn, end_pfn] = Initialized and sent to buddy

Hot-remove:

We need to be careful when removing memory, as adding and removing memory needs to be done with the same granularity. To check that this assumption is not violated, we check the memory range we want to remove and if a) any memory block has vmemmap pages and b) the range spans more than a single memory block, we scream out loud and refuse to proceed.

Quoted 5 lines

If all is good and the range was using memmap on memory (aka vmemmap pages), we construct an altmap structure so free_hugepage_table does the right thing and calls vmem_altmap_free instead of free_pagetable.

IIRC, there is a conflict with the hpage vmemmap freeing patch set, right? How are we going to handle that?

[...]

include/linux/memory.h struct memory_block { · line 37

Quoted 13 lines
include/linux/memory.h +5 -0
index 439a89e758d8..f09d60ef8ad9 100644
@@ -30,6 +30,11 @@ struct memory_block {
 	int phys_device;		/* to which fru does this belong? */
 	struct device dev;
 	int nid;			/* NID for this memory block */
+	unsigned long nr_vmemmap_pages;	/*
+					 * Number of vmemmap pages. These pages
+					 * lay at the beginning of the memory
+					 * block.
+					 */

I suggest moving this comment above nr_vmemmap_pages.

Quoted 132 lines
include/linux/memory.h +2 -1
 };
 
 int arch_get_memory_phys_device(unsigned long start_pfn);
@@ -81,7 +86,8 @@ static inline int memory_notify(unsigned long val, void *v)
 #else
 extern int register_memory_notifier(struct notifier_block *nb);
 extern void unregister_memory_notifier(struct notifier_block *nb);
-int create_memory_block_devices(unsigned long start, unsigned long size);
+int create_memory_block_devices(unsigned long start, unsigned long size,
+				unsigned long vmemmap_pages);
 void remove_memory_block_devices(unsigned long start, unsigned long size);
 extern void memory_dev_init(void);
 extern int memory_notify(unsigned long val, void *v);
include/linux/memory_hotplug.h +17 -4
index 15acce5ab106..67ea4a0b25bd 100644
@@ -70,6 +70,14 @@ typedef int __bitwise mhp_t;
  */
 #define MEMHP_MERGE_RESOURCE	((__force mhp_t)BIT(0))
 
+/*
+ * We want memmap (struct page array) to be self contained.
+ * To do so, we will use the beginning of the hot-added range to build
+ * the page tables for the memmap array that describes the entire range.
+ * Only selected architectures support it with SPARSE_VMEMMAP.
+ */
+#define MHP_MEMMAP_ON_MEMORY   ((__force mhp_t)BIT(1))
+
 /*
  * Extended parameters for memory hotplug:
  * altmap: alternative allocator for memmap array (optional)
@@ -113,11 +121,13 @@ extern int zone_grow_waitqueues(struct zone *zone, unsigned long nr_pages);
 extern int add_one_highpage(struct page *page, int pfn, int bad_ppro);
 /* VM interface that may be used by firmware interface */
 extern int online_pages(unsigned long pfn, unsigned long nr_pages,
-			int online_type, int nid);
+			unsigned long nr_vmemmap_pages, int online_type,
+			int nid);
 extern struct zone *test_pages_in_a_zone(unsigned long start_pfn,
 					 unsigned long end_pfn);
 extern void __offline_isolated_pages(unsigned long start_pfn,
-				     unsigned long end_pfn);
+				     unsigned long end_pfn,
+				     unsigned long buddy_start_pfn);
 
 typedef void (*online_page_callback_t)(struct page *page, unsigned int order);
 
@@ -312,7 +322,8 @@ static inline void pgdat_resize_init(struct pglist_data *pgdat) {}
 #ifdef CONFIG_MEMORY_HOTREMOVE
 
 extern void try_offline_node(int nid);
-extern int offline_pages(unsigned long start_pfn, unsigned long nr_pages);
+extern int offline_pages(unsigned long start_pfn, unsigned long nr_pages,
+			 unsigned long nr_vmemmap_pages);
 extern int remove_memory(int nid, u64 start, u64 size);
 extern void __remove_memory(int nid, u64 start, u64 size);
 extern int offline_and_remove_memory(int nid, u64 start, u64 size);
@@ -320,7 +331,8 @@ extern int offline_and_remove_memory(int nid, u64 start, u64 size);
 #else
 static inline void try_offline_node(int nid) {}
 
-static inline int offline_pages(unsigned long start_pfn, unsigned long nr_pages)
+static inline int offline_pages(unsigned long start_pfn, unsigned long nr_pages,
+				unsigned long nr_vmemmap_pages)
 {
 	return -EINVAL;
 }
@@ -364,6 +376,7 @@ extern struct zone *zone_for_pfn_range(int online_type, int nid, unsigned start_
 extern int arch_create_linear_mapping(int nid, u64 start, u64 size,
 				      struct mhp_params *params);
 void arch_remove_linear_mapping(u64 start, u64 size);
+extern bool mhp_supports_memmap_on_memory(unsigned long size);
 #endif /* CONFIG_MEMORY_HOTPLUG */
 
 #endif /* __LINUX_MEMORY_HOTPLUG_H */
include/linux/memremap.h +1 -1
index 86c6c368ce9b..3de5d482ac1a 100644
@@ -17,7 +17,7 @@ struct device;
  * @alloc: track pages consumed, private to vmemmap_populate()
  */
 struct vmem_altmap {
-	const unsigned long base_pfn;
+	unsigned long base_pfn;
 	const unsigned long end_pfn;
 	const unsigned long reserve;
 	unsigned long free;
include/linux/mmzone.h +5 -0
index b593316bff3d..d16b4411e422 100644
@@ -407,6 +407,11 @@ enum zone_type {
 	 *    techniques might use alloc_contig_range() to hide previously
 	 *    exposed pages from the buddy again (e.g., to implement some sort
 	 *    of memory unplug in virtio-mem).
+	 * 6. Memory-hotplug: when using memmap_on_memory and onlining the memory
+	 *    to the MOVABLE zone, the vmemmap pages are also placed in such
+	 *    zone. Such pages cannot be really moved around as they are
+	 *    self-stored in the range, but they are treated as movable when
+	 *    the range that they describe is about to be offlined.
 	 *
 	 * In general, no unmovable allocations that degrade memory offlining
 	 * should end up in ZONE_MOVABLE. Allocators (like alloc_contig_range())
mm/memory_hotplug.c +8 -2
index a8cef4955907..9371e7d3f583 100644
@@ -42,6 +42,8 @@
 #include "internal.h"
 #include "shuffle.h"
 
+static bool memmap_on_memory_enabled;
+
 /*
  * online_page_callback contains pointer to current page onlining function.
  * Initially it is generic_online_page(). If it is required it could be
@@ -606,10 +608,23 @@ void generic_online_page(struct page *page, unsigned int order)
 }
 EXPORT_SYMBOL_GPL(generic_online_page);
 
-static void online_pages_range(unsigned long start_pfn, unsigned long nr_pages)
+static void online_pages_range(unsigned long start_pfn, unsigned long nr_pages,
+			       unsigned long buddy_start_pfn)
 {
 	const unsigned long end_pfn = start_pfn + nr_pages;
-	unsigned long pfn;
+	unsigned long pfn = buddy_start_pfn;
+	unsigned int order = MAX_ORDER - 1;
+
+	/*

mm/memory_hotplug.c void generic_online_page(struct page *page, unsigned int order) · line 628

Quoted 10 lines
mm/memory_hotplug.c +9 -0
+	 * When using memmap_on_memory, the range might be unaligned as the
+	 * first pfns are used for vmemmap pages. Align it in case we need to.
+	 */
+	if (pfn & ((1 << order) - 1)) {
+		while (pfn & ((1 << order) - 1))
+			order--;
+		(*online_page_callback)(pfn_to_page(pfn), order);
+		pfn += 1 << order;
+	}
 

Most probably we can simplify once we assure buddy_start_pfn follows specific requirements (e.g., at least aligned to pageblock_nr_pages, see below).

Quoted 28 lines
mm/memory_hotplug.c +4 -3
 	/*
 	 * Online the pages in MAX_ORDER - 1 aligned chunks. The callback might
@@ -617,7 +632,7 @@ static void online_pages_range(unsigned long start_pfn, unsigned long nr_pages)
 	 * later). We account all pages as being online and belonging to this
 	 * zone ("present").
 	 */
-	for (pfn = start_pfn; pfn < end_pfn; pfn += MAX_ORDER_NR_PAGES)
+	for (; pfn < end_pfn; pfn += MAX_ORDER_NR_PAGES)
 		(*online_page_callback)(pfn_to_page(pfn), MAX_ORDER - 1);
 
 	/* mark all involved sections as online */
@@ -777,9 +792,9 @@ struct zone * zone_for_pfn_range(int online_type, int nid, unsigned start_pfn,
 }
 
 int __ref online_pages(unsigned long pfn, unsigned long nr_pages,
-		       int online_type, int nid)
+		       unsigned long nr_vmemmap_pages, int online_type, int nid)
 {
-	unsigned long flags;
+	unsigned long flags, buddy_start_pfn, buddy_nr_pages;
 	struct zone *zone;
 	int need_zonelists_rebuild = 0;
 	int ret;
@@ -790,11 +805,18 @@ int __ref online_pages(unsigned long pfn, unsigned long nr_pages,
 			 !IS_ALIGNED(pfn | nr_pages, PAGES_PER_SECTION)))
 		return -EINVAL;
 
+	buddy_start_pfn = pfn + nr_vmemmap_pages;

mm/memory_hotplug.c int __ref online_pages(unsigned long pfn, unsigned long nr_pages, · line 817

Quoted 10 lines
mm/memory_hotplug.c +5 -1
+	buddy_nr_pages = nr_pages - nr_vmemmap_pages;
+
 	mem_hotplug_begin();
 
 	/* associate pfn range with the zone */
 	zone = zone_for_pfn_range(online_type, nid, pfn, nr_pages);
-	move_pfn_range_to_zone(zone, pfn, nr_pages, NULL, MIGRATE_ISOLATE);
+	if (nr_vmemmap_pages)
+		move_pfn_range_to_zone(zone, pfn, nr_vmemmap_pages, NULL,
+				       MIGRATE_UNMOVABLE);

One sign that we have to take care that nr_vmemmap_pages has a certain minimum alignment.

mm/memory_hotplug.c int __ref online_pages(unsigned long pfn, unsigned long nr_pages, · line 835

Quoted 11 lines
mm/memory_hotplug.c +3 -1
+	move_pfn_range_to_zone(zone, buddy_start_pfn, buddy_nr_pages, NULL,
+			       MIGRATE_ISOLATE);
 
 	arg.start_pfn = pfn;
 	arg.nr_pages = nr_pages;
@@ -810,7 +832,7 @@ int __ref online_pages(unsigned long pfn, unsigned long nr_pages,
 	 * onlining, such that undo_isolate_page_range() works correctly.
 	 */
 	spin_lock_irqsave(&zone->lock, flags);
-	zone->nr_isolate_pageblock += nr_pages / pageblock_nr_pages;
+	zone->nr_isolate_pageblock += buddy_nr_pages / pageblock_nr_pages;

And, another sign :)

Quoted 13 lines
mm/memory_hotplug.c +1 -1
 	spin_unlock_irqrestore(&zone->lock, flags);
 
 	/*
@@ -823,7 +845,7 @@ int __ref online_pages(unsigned long pfn, unsigned long nr_pages,
 		setup_zone_pageset(zone);
 	}
 
-	online_pages_range(pfn, nr_pages);
+	online_pages_range(pfn, nr_pages, buddy_start_pfn);
 	zone->present_pages += nr_pages;
 
 	pgdat_resize_lock(zone->zone_pgdat, &flags);

mm/memory_hotplug.c int __ref online_pages(unsigned long pfn, unsigned long nr_pages, · line 863

Quoted 7 lines
mm/memory_hotplug.c +3 -1
@@ -836,7 +858,9 @@ int __ref online_pages(unsigned long pfn, unsigned long nr_pages,
 	zone_pcp_update(zone);
 
 	/* Basic onlining is complete, allow allocation of onlined pages. */
-	undo_isolate_page_range(pfn, pfn + nr_pages, MIGRATE_MOVABLE);
+	undo_isolate_page_range(buddy_start_pfn,
+				buddy_start_pfn + buddy_nr_pages,
+				MIGRATE_MOVABLE);

Yet another sign :)

Quoted 25 lines
mm/memory_hotplug.c +7 -0
 	/*
 	 * Freshly onlined pages aren't shuffled (e.g., all pages are placed to
@@ -1011,6 +1035,12 @@ static int online_memory_block(struct memory_block *mem, void *arg)
 	return device_online(&mem->dev);
 }
 
+bool mhp_supports_memmap_on_memory(unsigned long size)
+{
+	return memmap_on_memory_enabled &&
+	       size == memory_block_size_bytes();
+}
+
 /*
  * NOTE: The caller must call lock_device_hotplug() to serialize hotplug
  * and online/offline operations (triggered e.g. by sysfs).
@@ -1020,6 +1050,7 @@ static int online_memory_block(struct memory_block *mem, void *arg)
 int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags)
 {
 	struct mhp_params params = { .pgprot = PAGE_KERNEL };
+	struct vmem_altmap mhp_altmap = {};
 	u64 start, size;
 	bool new_node = false;
 	int ret;

mm/memory_hotplug.c int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags) · line 1083

Quoted 7 lines
mm/memory_hotplug.c +4 -0
@@ -1046,13 +1077,32 @@ int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags)
 		goto error;
 	new_node = ret;
 
+	/*
+	 * Return -EINVAL if the caller specified MHP_MEMMAP_ON_MEMORY and we do
+	 * not support it.
+	 */

The "Return -EINVAL" part is obvious. Maybe drop that comment.

mm/memory_hotplug.c int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags) · line 1095

Quoted 12 lines
mm/memory_hotplug.c +12 -0
+	if ((mhp_flags & MHP_MEMMAP_ON_MEMORY) &&
+	    !mhp_supports_memmap_on_memory(size)) {
+		ret = -EINVAL;
+		goto error;
+	}
+
+	/*
+	 * Self hosted memmap array
+	 */
+	if (mhp_flags & MHP_MEMMAP_ON_MEMORY) {
+		mhp_altmap.free = size >> PAGE_SHIFT;
+		mhp_altmap.base_pfn = start >> PAGE_SHIFT;

Could use PFN_PHYS() for both.

mm/memory_hotplug.c int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags) · line 1098

Quoted 3 lines
mm/memory_hotplug.c +3 -0
+		params.altmap = &mhp_altmap;
+	}
+

I'd combine both parts

if (mhp_flags & MHP_MEMMAP_ON_MEMORY) {
	if (!mhp_supports_memmap_on_memory(size)) {
		ret = -EINVAL;
		goto error;
	}
	mhp_altmap.free = PFN_PHYS(size);
	mhp_altmap.base_pfn = PFN_PHYS(size);
	params.altmap = &mhp_altmap;

}

mm/memory_hotplug.c int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags) · line 1105

Quoted 8 lines
mm/memory_hotplug.c +1 -1
 	/* call arch's memory hotadd */
 	ret = arch_add_memory(nid, start, size, &params);
 	if (ret < 0)
 		goto error;
 
 	/* create memory block devices after memory was added */
-	ret = create_memory_block_devices(start, size);
+	ret = create_memory_block_devices(start, size, mhp_altmap.alloc);

Interresting, so we automatically support differeing sizeof(struct page). I guess it will be problematic in case of sizeof(struct page) != 64, because then, we might not have multiples of 2MB for the memmap of a memory block.

IIRC, it could happen that if we add two consecutive memory blocks, that the second one might reuse parts of the vmemmap residing on the first memory block. If you remove the first one, you might be in trouble.

E.g., on x86-64
vmemmap_populate()->vmemmap_populate_hugepages()->vmemmap_alloc_block_buf():
- Populate a huge page

vmemmap_free()->remove_pagetable()...->remove_pmd_table():
- memchr_inv() will leave the hugepage populated.

Do we want to fence that off, maybe in mhp_supports_memmap_on_memory()? Or do we somehow want to fix that? We should never populate partial huge pages from an altmap ...

But maybe I am missing something.

Quoted 28 lines
mm/memory_hotplug.c +6 -2
 	if (ret) {
 		arch_remove_memory(nid, start, size, NULL);
 		goto error;
@@ -1448,10 +1498,11 @@ static int count_system_ram_pages_cb(unsigned long start_pfn,
 	return 0;
 }
 
-int __ref offline_pages(unsigned long start_pfn, unsigned long nr_pages)
+int __ref offline_pages(unsigned long start_pfn, unsigned long nr_pages,
+			unsigned long nr_vmemmap_pages)
 {
 	const unsigned long end_pfn = start_pfn + nr_pages;
-	unsigned long pfn, system_ram_pages = 0;
+	unsigned long pfn, buddy_start_pfn, buddy_nr_pages, system_ram_pages = 0;
 	unsigned long flags;
 	struct zone *zone;
 	struct memory_notify arg;
@@ -1463,6 +1514,9 @@ int __ref offline_pages(unsigned long start_pfn, unsigned long nr_pages)
 			 !IS_ALIGNED(start_pfn | nr_pages, PAGES_PER_SECTION)))
 		return -EINVAL;
 
+	buddy_start_pfn = start_pfn + nr_vmemmap_pages;
+	buddy_nr_pages = nr_pages - nr_vmemmap_pages;
+
 	mem_hotplug_begin();
 
 	/*

mm/memory_hotplug.c int __ref offline_pages(unsigned long start_pfn, unsigned long nr_pages) · line 1557

Quoted 7 lines
mm/memory_hotplug.c +1 -1
@@ -1498,7 +1552,7 @@ int __ref offline_pages(unsigned long start_pfn, unsigned long nr_pages)
 	zone_pcp_disable(zone);
 
 	/* set above range as isolated */
-	ret = start_isolate_page_range(start_pfn, end_pfn,
+	ret = start_isolate_page_range(buddy_start_pfn, end_pfn,
 				       MIGRATE_MOVABLE,
 				       MEMORY_OFFLINE | REPORT_FAILURE);

If buddy_start_pfn is not aligned to pageblocks, start_isolate_page_range will BUG_ON(). another sign that we need to check that somewhere while adding that nr_vmemmap_pages is suitable.

[...]

mm/memory_hotplug.c static int check_memblock_offlined_cb(struct memory_block *mem, void *arg) · line 1692

Quoted 2 lines
mm/memory_hotplug.c +1 -0
+static int get_memblock_vmemmap_pages_cb(struct memory_block *mem, void *arg)

Let's avoid using the "memblock" terminology. Just call it "get_nr_vmemmap_pages_cb" or maybe "detect_memmap_on_memory_cb"

mm/memory_hotplug.c static int check_memblock_offlined_cb(struct memory_block *mem, void *arg) · line 1699

Quoted 7 lines
mm/memory_hotplug.c +7 -0
+{
+	unsigned long *nr_vmemmap_pages = (unsigned long *)arg;
+	int ret = !mem->nr_vmemmap_pages;
+
+	if (!ret)
+		*nr_vmemmap_pages += mem->nr_vmemmap_pages;
+	return ret;

I think you can avoid "ret".

Also, we want to stop whenever we find "mem->nr_vmemmap_pages != 0", right? AFAIU, you would currently stop whenever you find mem->nr_vmemmap_pages == 0", which is wrong.

*nr_vmemmap_pages += mem->nr_vmemmap_pages;
/* Stop whenever we find any sign of MHP_MEMMAP_ON_MEMORY. */ return mem->nr_vmemmap_pages;

Quoted 20 lines
mm/memory_hotplug.c +6 -0
+}
+
 static int check_cpu_on_node(pg_data_t *pgdat)
 {
 	int cpu;
@@ -1709,6 +1773,9 @@ EXPORT_SYMBOL(try_offline_node);
 static int __ref try_remove_memory(int nid, u64 start, u64 size)
 {
 	int rc = 0;
+	struct vmem_altmap mhp_altmap = {};
+	struct vmem_altmap *altmap = NULL;
+	unsigned long nr_vmemmap_pages = 0;
 
 	BUG_ON(check_hotplug_memory_range(start, size));
 
@@ -1721,6 +1788,37 @@ static int __ref try_remove_memory(int nid, u64 start, u64 size)
 	if (rc)
 		return rc;
 
+	/*

mm/memory_hotplug.c static int __ref try_remove_memory(int nid, u64 start, u64 size) · line 1801

Quoted 10 lines
mm/memory_hotplug.c +10 -0
+	 * When using MHP_MEMMAP_ON_MEMORY, add_memory and remove_memory needs
+	 * to work with the same granularity.
+	 * Check that we are not violating that assumption.
+	 * The only way we have right now is to walk all the memory
+	 * blocks we are trying to remove and see if any has vmemmap pages.
+	 * We should not have any as memmap_on_memory only works with ranges
+	 * that spans a single memory block, so if we found any and size spans
+	 * more than a memory block, we know we have to scream.
+	 * We only need to check this if memmap_on_memory_enabled=true.
+	 */

Can we simplify that comment?

/*
* We only support remvoing memory added with MHP_MEMMAP_ON_MEMORY in
* the same granularity it was added - a single memory block. */

mm/memory_hotplug.c static int __ref try_remove_memory(int nid, u64 start, u64 size) · line 1804

Quoted 3 lines
mm/memory_hotplug.c +3 -0
+	if (memmap_on_memory_enabled) {
+		(void)walk_memory_blocks(start, size, &nr_vmemmap_pages,
+					 get_memblock_vmemmap_pages_cb);

you can drop the "(void)", just like other users that ignore the return value.

Quoted 17 lines
mm/memory_hotplug.c +17 -0
+		if (nr_vmemmap_pages) {
+			if (size != memory_block_size_bytes()) {
+				pr_warn("Refuse to remove %#llx - %#llx, wrong granularity\n"
+					 start, start + size);
+				return -EINVAL;
+			}
+
+			/*
+			 * Let remove_pmd_table->free_hugepage_table
+			 * do the right thing if we used vmem_altmap
+			 * when hot-adding the range.
+			 */
+			mhp_altmap.alloc = nr_vmemmap_pages;
+			altmap = &mhp_altmap;
+		}
+	}
+
Signature · 3 lines
Thanks,

David / dhildenb
author maintainer
#10 lore raw

Re: [PATCH 2/5] mm,memory_hotplug: Allocate memmap from the added memory range

Quoted 2 lines · David Hildenbrand

IIRC, there is a conflict with the hpage vmemmap freeing patch set, right? How are we going to handle that?

First of all, sorry for the lateness David, was a bit busy.

AFAIK, there is no conflict at the moment.
hpage vmemmap populates with basepages (PMD support is disabled).

But even with PMD support enabled, I cannot see how it can interfere.

From the mhp point of view vmemmap pages only matter when allocating it and when freeing it.
The only moment we would have to worry about is when we have to get rid of them (aka: the memory range that describe is being hot-removed), but the hpage vmemmap patchset makes sure that by the time we return from dissolve_free_huge_page the vemmap pages look as they did before the remapping.

But as I said, we do not have to worry about this one as that feature only works when populating with base pages.

Quoted 7 lines
Quoted 7 lines
include/linux/memory.h +0 -0
index 439a89e758d8..f09d60ef8ad9 100644
@@ -30,6 +30,11 @@ struct memory_block {
 	int phys_device;		/* to which fru does this belong? */
 	struct device dev;
Quoted 8 lines

include/linux/memory.h struct memory_block { · line 37

Quoted 6 lines
include/linux/memory.h +5 -0
 	int nid;			/* NID for this memory block */
+	unsigned long nr_vmemmap_pages;	/*
+					 * Number of vmemmap pages. These pages
+					 * lay at the beginning of the memory
+					 * block.
+					 */

I suggest moving this comment above nr_vmemmap_pages.

Sure, will do.

Quoted 16 lines
Quoted 16 lines
include/linux/memory.h +12 -2
-static void online_pages_range(unsigned long start_pfn, unsigned long nr_pages)
+static void online_pages_range(unsigned long start_pfn, unsigned long nr_pages,
+			       unsigned long buddy_start_pfn)
 {
 	const unsigned long end_pfn = start_pfn + nr_pages;
-	unsigned long pfn;
+	unsigned long pfn = buddy_start_pfn;
+	unsigned int order = MAX_ORDER - 1;
+
+	/*
+	 * When using memmap_on_memory, the range might be unaligned as the
+	 * first pfns are used for vmemmap pages. Align it in case we need to.
+	 */
+	if (pfn & ((1 << order) - 1)) {
+		while (pfn & ((1 << order) - 1))
+			order--;
Quoted 8 lines

include/linux/memory.h

Quoted 4 lines
include/linux/memory.h +3 -0
+		(*online_page_callback)(pfn_to_page(pfn), order);
+		pfn += 1 << order;
+	}

Most probably we can simplify once we assure buddy_start_pfn follows specific requirements (e.g., at least aligned to pageblock_nr_pages, see below).

Yeah, on x86_64 this is pageblock_nr_pages aligned, but will have to check on arm64.

Quoted 5 lines
Quoted 5 lines
include/linux/memory.h +3 -0
+	buddy_start_pfn = pfn + nr_vmemmap_pages;
+	buddy_nr_pages = nr_pages - nr_vmemmap_pages;
+
 	mem_hotplug_begin();
Quoted 9 lines

include/linux/memory.h

Quoted 6 lines
include/linux/memory.h +3 -1
 	/* associate pfn range with the zone */
 	zone = zone_for_pfn_range(online_type, nid, pfn, nr_pages);
-	move_pfn_range_to_zone(zone, pfn, nr_pages, NULL, MIGRATE_ISOLATE);
+	if (nr_vmemmap_pages)
+		move_pfn_range_to_zone(zone, pfn, nr_vmemmap_pages, NULL,
+				       MIGRATE_UNMOVABLE);

One sign that we have to take care that nr_vmemmap_pages has a certain minimum alignment.

Right.

Quoted 13 lines

include/linux/memory.h int __ref online_pages(unsigned long pfn, unsigned long nr_pages, · line 835

Quoted 11 lines
include/linux/memory.h +3 -1
+	move_pfn_range_to_zone(zone, buddy_start_pfn, buddy_nr_pages, NULL,
+			       MIGRATE_ISOLATE);
 
 	arg.start_pfn = pfn;
 	arg.nr_pages = nr_pages;
@@ -810,7 +832,7 @@ int __ref online_pages(unsigned long pfn, unsigned long nr_pages,
 	 * onlining, such that undo_isolate_page_range() works correctly.
 	 */
 	spin_lock_irqsave(&zone->lock, flags);
-	zone->nr_isolate_pageblock += nr_pages / pageblock_nr_pages;
+	zone->nr_isolate_pageblock += buddy_nr_pages / pageblock_nr_pages;

And, another sign :)

Yap

Quoted 8 lines
Quoted 8 lines
include/linux/memory.h +1 -1
-	online_pages_range(pfn, nr_pages);
+	online_pages_range(pfn, nr_pages, buddy_start_pfn);
 	zone->present_pages += nr_pages;
 
include/linux/memory.h +0 -0
 	pgdat_resize_lock(zone->zone_pgdat, &flags);
@@ -836,7 +858,9 @@ int __ref online_pages(unsigned long pfn, unsigned long nr_pages,
 	zone_pcp_update(zone);
 
Quoted 7 lines

include/linux/memory.h int __ref online_pages(unsigned long pfn, unsigned long nr_pages, · line 863

Quoted 5 lines
include/linux/memory.h +3 -1
 	/* Basic onlining is complete, allow allocation of onlined pages. */
-	undo_isolate_page_range(pfn, pfn + nr_pages, MIGRATE_MOVABLE);
+	undo_isolate_page_range(buddy_start_pfn,
+				buddy_start_pfn + buddy_nr_pages,
+				MIGRATE_MOVABLE);

Yet another sign :)

Yap

Quoted 11 lines
Quoted 11 lines
include/linux/memory.h +1 -0
 int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags)
 {
 	struct mhp_params params = { .pgprot = PAGE_KERNEL };
+	struct vmem_altmap mhp_altmap = {};
 	u64 start, size;
 	bool new_node = false;
 	int ret;
@@ -1046,13 +1077,32 @@ int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags)
 		goto error;
 	new_node = ret;
 
Quoted 6 lines

include/linux/memory.h int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags) · line 1083

Quoted 4 lines
include/linux/memory.h +4 -0
+	/*
+	 * Return -EINVAL if the caller specified MHP_MEMMAP_ON_MEMORY and we do
+	 * not support it.
+	 */

The "Return -EINVAL" part is obvious. Maybe drop that comment.

Sure.

Quoted 5 lines

include/linux/memory.h int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags) · line 1086

Quoted 3 lines
include/linux/memory.h +3 -0
+	if (mhp_flags & MHP_MEMMAP_ON_MEMORY) {
+		mhp_altmap.free = size >> PAGE_SHIFT;
+		mhp_altmap.base_pfn = start >> PAGE_SHIFT;

Could use PFN_PHYS() for both.

Will do.

Quoted 11 lines

I'd combine both parts

if (mhp_flags & MHP_MEMMAP_ON_MEMORY) {
	if (!mhp_supports_memmap_on_memory(size)) {
		ret = -EINVAL;
		goto error;
	}
	mhp_altmap.free = PFN_PHYS(size);
	mhp_altmap.base_pfn = PFN_PHYS(size);
	params.altmap = &mhp_altmap;
Quoted 1 line

}

Ok, cleaner.

Quoted 13 lines

Interresting, so we automatically support differeing sizeof(struct page). I guess it will be problematic in case of sizeof(struct page) != 64, because then, we might not have multiples of 2MB for the memmap of a memory block.

IIRC, it could happen that if we add two consecutive memory blocks, that the second one might reuse parts of the vmemmap residing on the first memory block. If you remove the first one, you might be in trouble.

E.g., on x86-64
vmemmap_populate()->vmemmap_populate_hugepages()->vmemmap_alloc_block_buf():
- Populate a huge page

Quoted 8 lines

vmemmap_free()->remove_pagetable()...->remove_pmd_table():
- memchr_inv() will leave the hugepage populated.

Do we want to fence that off, maybe in mhp_supports_memmap_on_memory()? Or do we somehow want to fix that? We should never populate partial huge pages from an altmap ...

But maybe I am missing something.

No, you are not missing anything.

I think that remove_pmd_table() should be fixed. vmemmap_populate_hugepages() allocates PMD_SIZE chunk and marks the PMD as PAGE_KERNEL_LARGE, so when remove_pmd_table() sees that 1) the PMD is large and 2) the chunk is not aligned, the memset() should be writing PAGE_INUSE for a PMD chunk.

I do not really a see a reason why this should not happen. Actually, we will be leaving pagetables behind as we never get to free the PMD chunk when sizeof(struct page) is not a multiple of 2MB.

I plan to fix remove_pmd_table(), but for now let us not allow to use this feature if the size of a struct page is not 64.
Let us keep it simply for the time being, shall we?

Quoted 9 lines

include/linux/memory.h int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags) · line 1090

Quoted 5 lines
include/linux/memory.h +1 -1
 	/* set above range as isolated */
-	ret = start_isolate_page_range(start_pfn, end_pfn,
+	ret = start_isolate_page_range(buddy_start_pfn, end_pfn,
 				       MIGRATE_MOVABLE,
 				       MEMORY_OFFLINE | REPORT_FAILURE);

If buddy_start_pfn is not aligned to pageblocks, start_isolate_page_range will BUG_ON(). another sign that we need to check that somewhere while adding that nr_vmemmap_pages is suitable.

Will come up with a way to 1) ensure that vmemmap pages is aligned and
2) disallow to use the feature in case it cannot be aligned for some reason.

Quoted 2 lines

Let's avoid using the "memblock" terminology. Just call it "get_nr_vmemmap_pages_cb" or maybe "detect_memmap_on_memory_cb"

Fine.

Quoted 9 lines

include/linux/memory.h int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags) · line 1096

Quoted 6 lines
include/linux/memory.h +6 -0
+	unsigned long *nr_vmemmap_pages = (unsigned long *)arg;
+	int ret = !mem->nr_vmemmap_pages;
+
+	if (!ret)
+		*nr_vmemmap_pages += mem->nr_vmemmap_pages;
+	return ret;

I think you can avoid "ret".

Quoted 7 lines

Also, we want to stop whenever we find "mem->nr_vmemmap_pages != 0", right? AFAIU, you would currently stop whenever you find mem->nr_vmemmap_pages == 0", which is wrong.

*nr_vmemmap_pages += mem->nr_vmemmap_pages;
/* Stop whenever we find any sign of MHP_MEMMAP_ON_MEMORY. */ return mem->nr_vmemmap_pages;

Yeah, true is that current logic is wrong, but we do not want to stop when mem->nr_vmemmap_pages == 0 either.
Since we want to avoid people to remove with a different granularity than it was added, we need to check the whole memory range that is going to be hot-removed and check for mem->nr_vmemmap_pages to make sure that the range does not have any self-hosted vmemmap pages.

E.g:

[memory_block not using vmemmap] memblock#0
[memory block using vmemmap] memblock#1

So, if someone tries to remove memblock#0 and memblock#1 and once, we need to check whether any of the memblocks used self-hosted vmemmap, and if so, scream out loud if we are trying to remove site != memblock_size.

Quoted 14 lines

include/linux/memory.h int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags) · line 1107

Quoted 11 lines
include/linux/memory.h +11 -0
+	/*
+	 * When using MHP_MEMMAP_ON_MEMORY, add_memory and remove_memory needs
+	 * to work with the same granularity.
+	 * Check that we are not violating that assumption.
+	 * The only way we have right now is to walk all the memory
+	 * blocks we are trying to remove and see if any has vmemmap pages.
+	 * We should not have any as memmap_on_memory only works with ranges
+	 * that spans a single memory block, so if we found any and size spans
+	 * more than a memory block, we know we have to scream.
+	 * We only need to check this if memmap_on_memory_enabled=true.
+	 */

Can we simplify that comment?

Quoted 4 lines

/*
* We only support remvoing memory added with MHP_MEMMAP_ON_MEMORY in
* the same granularity it was added - a single memory block. */

Yeah.

Quoted 7 lines

include/linux/memory.h int __ref add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags)

Quoted 4 lines
include/linux/memory.h +3 -0
+	if (memmap_on_memory_enabled) {
+		(void)walk_memory_blocks(start, size, &nr_vmemmap_pages,
+					 get_memblock_vmemmap_pages_cb);

you can drop the "(void)", just like other users that ignore the return value.

ok

Thanks for the throughout review!

Signature · 2 lines
Oscar Salvador
SUSE L3
author maintainer
#11 lore raw

Re: [PATCH 2/5] mm,memory_hotplug: Allocate memmap from the added memory range

Quoted 30 lines · Oscar Salvador
Quoted 13 lines

Interresting, so we automatically support differeing sizeof(struct page). I guess it will be problematic in case of sizeof(struct page) != 64, because then, we might not have multiples of 2MB for the memmap of a memory block.

IIRC, it could happen that if we add two consecutive memory blocks, that the second one might reuse parts of the vmemmap residing on the first memory block. If you remove the first one, you might be in trouble.

E.g., on x86-64
vmemmap_populate()->vmemmap_populate_hugepages()->vmemmap_alloc_block_buf():
- Populate a huge page

Quoted 8 lines

vmemmap_free()->remove_pagetable()...->remove_pmd_table():
- memchr_inv() will leave the hugepage populated.

Do we want to fence that off, maybe in mhp_supports_memmap_on_memory()? Or do we somehow want to fix that? We should never populate partial huge pages from an altmap ...

But maybe I am missing something.

No, you are not missing anything.

I think that remove_pmd_table() should be fixed. vmemmap_populate_hugepages() allocates PMD_SIZE chunk and marks the PMD as PAGE_KERNEL_LARGE, so when remove_pmd_table() sees that 1) the PMD is large and 2) the chunk is not aligned, the memset() should be writing PAGE_INUSE for a PMD chunk.

Quoted 7 lines

I do not really a see a reason why this should not happen. Actually, we will be leaving pagetables behind as we never get to free the PMD chunk when sizeof(struct page) is not a multiple of 2MB.

I plan to fix remove_pmd_table(), but for now let us not allow to use this feature if the size of a struct page is not 64.
Let us keep it simply for the time being, shall we?

My first intention was:

arch/x86/mm/init_64.c +7 -25
index b5a3fa4033d3..0c9756a2eb24 100644
@@ -1044,32 +1044,14 @@ remove_pmd_table(pmd_t *pmd_start, unsigned long addr, unsigned long end,
 			continue;
 
 		if (pmd_large(*pmd)) {
-			if (IS_ALIGNED(addr, PMD_SIZE) &&
-			    IS_ALIGNED(next, PMD_SIZE)) {
-				if (!direct)
-					free_hugepage_table(pmd_page(*pmd),
-							    altmap);
-
-				spin_lock(&init_mm.page_table_lock);
-				pmd_clear(pmd);
-				spin_unlock(&init_mm.page_table_lock);
-				pages++;
-			} else {
-				/* If here, we are freeing vmemmap pages. */
-				memset((void *)addr, PAGE_INUSE, next - addr);
-
-				page_addr = page_address(pmd_page(*pmd));
-				if (!memchr_inv(page_addr, PAGE_INUSE,
-						PMD_SIZE)) {
-					free_hugepage_table(pmd_page(*pmd),
-							    altmap);
-
-					spin_lock(&init_mm.page_table_lock);
-					pmd_clear(pmd);
-					spin_unlock(&init_mm.page_table_lock);
-				}
-			}
+			if (!direct)
+				free_hugepage_table(pmd_page(*pmd),
+						    altmap);
 
+			spin_lock(&init_mm.page_table_lock);
+			pmd_clear(pmd);
+			spin_unlock(&init_mm.page_table_lock);
+			pages++;
 			continue;
 		}
 

I did not try it out yet and this might explode badly, but AFAICS, a PMD size chunk is always allocated even when the section does not spand 2MB. E.g: sizeof(struct page) = 56.

PAGES_PER_SECTION * 64 = 2097152
PAGES_PER_SECTION * 56 = 1835008

Even in the latter case, vmemmap_populate_hugepages will allocate a PMD size chunk to satisfy the unaligned range.
So, unless I am missing something, it should not be a problem to free that 2MB in remove_pmd_table when 1) the PMD is large and 2) the range is not aligned.

Signature · 2 lines
Oscar Salvador
SUSE L3
#12 lore raw

Re: [PATCH 2/5] mm,memory_hotplug: Allocate memmap from the added memory range

Quoted 89 lines · Oscar Salvador
Quoted 30 lines · Oscar Salvador
Quoted 13 lines

Interresting, so we automatically support differeing sizeof(struct page). I guess it will be problematic in case of sizeof(struct page) != 64, because then, we might not have multiples of 2MB for the memmap of a memory block.

IIRC, it could happen that if we add two consecutive memory blocks, that the second one might reuse parts of the vmemmap residing on the first memory block. If you remove the first one, you might be in trouble.

E.g., on x86-64
vmemmap_populate()->vmemmap_populate_hugepages()->vmemmap_alloc_block_buf():
- Populate a huge page

Quoted 8 lines

vmemmap_free()->remove_pagetable()...->remove_pmd_table():
- memchr_inv() will leave the hugepage populated.

Do we want to fence that off, maybe in mhp_supports_memmap_on_memory()? Or do we somehow want to fix that? We should never populate partial huge pages from an altmap ...

But maybe I am missing something.

No, you are not missing anything.

I think that remove_pmd_table() should be fixed. vmemmap_populate_hugepages() allocates PMD_SIZE chunk and marks the PMD as PAGE_KERNEL_LARGE, so when remove_pmd_table() sees that 1) the PMD is large and 2) the chunk is not aligned, the memset() should be writing PAGE_INUSE for a PMD chunk.

Quoted 7 lines

I do not really a see a reason why this should not happen. Actually, we will be leaving pagetables behind as we never get to free the PMD chunk when sizeof(struct page) is not a multiple of 2MB.

I plan to fix remove_pmd_table(), but for now let us not allow to use this feature if the size of a struct page is not 64.
Let us keep it simply for the time being, shall we?

My first intention was:

arch/x86/mm/init_64.c +7 -25
index b5a3fa4033d3..0c9756a2eb24 100644
@@ -1044,32 +1044,14 @@ remove_pmd_table(pmd_t *pmd_start, unsigned long addr, unsigned long end,
 			continue;
 
 		if (pmd_large(*pmd)) {
-			if (IS_ALIGNED(addr, PMD_SIZE) &&
-			    IS_ALIGNED(next, PMD_SIZE)) {
-				if (!direct)
-					free_hugepage_table(pmd_page(*pmd),
-							    altmap);
-
-				spin_lock(&init_mm.page_table_lock);
-				pmd_clear(pmd);
-				spin_unlock(&init_mm.page_table_lock);
-				pages++;
-			} else {
-				/* If here, we are freeing vmemmap pages. */
-				memset((void *)addr, PAGE_INUSE, next - addr);
-
-				page_addr = page_address(pmd_page(*pmd));
-				if (!memchr_inv(page_addr, PAGE_INUSE,
-						PMD_SIZE)) {
-					free_hugepage_table(pmd_page(*pmd),
-							    altmap);
-
-					spin_lock(&init_mm.page_table_lock);
-					pmd_clear(pmd);
-					spin_unlock(&init_mm.page_table_lock);
-				}
-			}
+			if (!direct)
+				free_hugepage_table(pmd_page(*pmd),
+						    altmap);
 
+			spin_lock(&init_mm.page_table_lock);
+			pmd_clear(pmd);
+			spin_unlock(&init_mm.page_table_lock);
+			pages++;
 			continue;
 		}
 

I did not try it out yet and this might explode badly, but AFAICS, a PMD size chunk is always allocated even when the section does not spand 2MB. E.g: sizeof(struct page) = 56.

Quoted 7 lines

PAGES_PER_SECTION * 64 = 2097152
PAGES_PER_SECTION * 56 = 1835008

Even in the latter case, vmemmap_populate_hugepages will allocate a PMD size chunk to satisfy the unaligned range.
So, unless I am missing something, it should not be a problem to free that 2MB in remove_pmd_table when 1) the PMD is large and 2) the range is not aligned.

Assume you have two consecutive memory blocks with 56 sizeof(struct page). The first one allocates a PMD (2097152) but only consumes 1835008, the second one reuses the remaining part and allocates another PMD (1835008), only using parts of it.

Ripping out a memory block, along with the PMD in the vmemmap would remove parts of the vmemmap of another memory block.

You might want to take a look at:

commit 2c114df071935762ffa88144cdab03d84beaa702

Author: David Hildenbrand <[email protected]>
Date:   Wed Jul 22 11:45:58 2020 +0200

s390/vmemmap: avoid memset(PAGE_UNUSED) when adding consecutive sections

Let's avoid memset(PAGE_UNUSED) when adding consecutive sections, whereby the vmemmap of a single section does not span full PMDs.

Cc Vasily Gorbik <[email protected]>
Cc Christian Borntraeger <[email protected]>
Cc Gerald Schaefer <[email protected]>

Message-Id: <[email protected]>

commit cd5781d63eaf6dbf89532d8c7c214786b767ee16

Author: David Hildenbrand <[email protected]>
Date:   Wed Jul 22 11:45:57 2020 +0200

s390/vmemmap: remember unused sub-pmd ranges

With a memmap size of 56 bytes or 72 bytes per page, the memmap for a 256 MB section won't span full PMDs. As we populate single sections and depopulate single sections, the depopulation step would not be able to free all vmemmap pmds anymore.

Do it similarly to x86, marking the unused memmap ranges in a special way (pad it with 0xFD).

This allows us to add/remove sections, cleaning up all allocated vmemmap pages even if the memmap size is not multiple of 16 bytes per page.

A 56 byte memmap can, for example, be created with !CONFIG_MEMCG and !CONFIG_SLUB.

Cc Vasily Gorbik <[email protected]>
Cc Christian Borntraeger <[email protected]>
Cc Gerald Schaefer <[email protected]>

Message-Id: <[email protected]>

But again, the root issue I see is that we can play such games with anonymous pages, but not with memory residing in the altmap space.

Signature · 3 lines
Thanks,

David / dhildenb
author maintainer
#13 lore raw

Re: [PATCH 2/5] mm,memory_hotplug: Allocate memmap from the added memory range

Quoted 7 lines · David Hildenbrand

Assume you have two consecutive memory blocks with 56 sizeof(struct page). The first one allocates a PMD (2097152) but only consumes 1835008, the second one reuses the remaining part and allocates another PMD (1835008), only using parts of it.

Ripping out a memory block, along with the PMD in the vmemmap would remove parts of the vmemmap of another memory block.

Bleh, yeah, I was confused, you are right.

Quoted 1 line

You might want to take a look at:

Thanks a lot for the hints, I will have a look ;-)

Signature · 2 lines
Oscar Salvador
SUSE L3
#14 lore raw

Re: [PATCH 2/5] mm,memory_hotplug: Allocate memmap from the added memory range

In reply to Oscar Salvador
Quoted 8 lines · Oscar Salvador
Quoted 2 lines · David Hildenbrand

IIRC, there is a conflict with the hpage vmemmap freeing patch set, right? How are we going to handle that?

First of all, sorry for the lateness David, was a bit busy.

AFAIK, there is no conflict at the moment.
hpage vmemmap populates with basepages (PMD support is disabled).

I don't see how that makes a difference? We end up consuming the altmap via vmemmap_alloc_block_buf(). Read below, most probably I am missing something.

Quoted 9 lines

But even with PMD support enabled, I cannot see how it can interfere.

From the mhp point of view vmemmap pages only matter when allocating it and when freeing it.
The only moment we would have to worry about is when we have to get rid of them (aka: the memory range that describe is being hot-removed), but the hpage vmemmap patchset makes sure that by the time we return from dissolve_free_huge_page the vemmap pages look as they did before the remapping.

Aehm I don't think that's the case. there might be a vmemmap but these are then other pages than the altmap.

Quoted 3 lines

But as I said, we do not have to worry about this one as that feature only works when populating with base pages.

I'm confused.

1. Assume we hotplug memory, online it to ZONE_MOVABLE. The vmemmap gets allocated from altmap space.

2. Allocate huge pages. The vmemmap gets freed, which ends up freeing pages in the altmap space that are even PageReserved() and not even accounted properly in managed pages. Problem?

3. Try offlining memory, skip the altmap space. The altmap pages might now be used for something else in the meantime. Problem?

This should hold with both, basepages and hugepages in the vmemmap. What am I missing?

[...]

Quoted 17 lines
Quoted 8 lines

IIRC, it could happen that if we add two consecutive memory blocks, that the second one might reuse parts of the vmemmap residing on the first memory block. If you remove the first one, you might be in trouble.

E.g., on x86-64
vmemmap_populate()->vmemmap_populate_hugepages()->vmemmap_alloc_block_buf():
- Populate a huge page

Quoted 8 lines

vmemmap_free()->remove_pagetable()...->remove_pmd_table():
- memchr_inv() will leave the hugepage populated.

Do we want to fence that off, maybe in mhp_supports_memmap_on_memory()? Or do we somehow want to fix that? We should never populate partial huge pages from an altmap ...

But maybe I am missing something.

Quoted 7 lines

No, you are not missing anything.

I think that remove_pmd_table() should be fixed. vmemmap_populate_hugepages() allocates PMD_SIZE chunk and marks the PMD as PAGE_KERNEL_LARGE, so when remove_pmd_table() sees that 1) the PMD is large and 2) the chunk is not aligned, the memset() should be writing PAGE_INUSE for a PMD chunk.

I tried to handle such stuff (dealing with partially present vmemmap on a PMD) a little bit smarter in the s390x code I added recently.

Anyhow, we should never ever consume memory from an altmap when it might cover more than the added range; that really only works with anon memory, but not with memory from the altmap space.

Quoted 8 lines

I do not really a see a reason why this should not happen. Actually, we will be leaving pagetables behind as we never get to free the PMD chunk when sizeof(struct page) is not a multiple of 2MB.

I plan to fix remove_pmd_table(), but for now let us not allow to use this feature if the size of a struct page is not 64.
Let us keep it simply for the time being, shall we?

Fine with me, as long as we never ever accidentally allocate more from the altmap than we actually need - the vmemmap must either span complete PMDs or we have to fallback to PTEs for the problematic parts.

So while we require "IS_ALIGNED(nr_vmemmap_pages, pageblock_nr_pages)", we actually require here "IS_ALIGNED(nr_vmemmap_pages, PMD_SIZE >> PAGE_SHIFT)" IIUC. (+ alignment of the start address)

For x86-64 this holds, for arm64 as well at least for 4k base pages. I *hope* 64k base pages never ever use PMDs (512MB) in the vmemmap when hotplugging :D (did not check, but I assume they don't, would be horribly ugly otherwise)

Quoted 33 lines
Quoted 10 lines

mm/memory_hotplug.c int __ref offline_pages(unsigned long start_pfn, unsigned long nr_pages) · line 1557

Quoted 6 lines
mm/memory_hotplug.c +1 -1
 	/* set above range as isolated */
-	ret = start_isolate_page_range(start_pfn, end_pfn,
+	ret = start_isolate_page_range(buddy_start_pfn, end_pfn,
 				       MIGRATE_MOVABLE,
 				       MEMORY_OFFLINE | REPORT_FAILURE);

If buddy_start_pfn is not aligned to pageblocks, start_isolate_page_range will BUG_ON(). another sign that we need to check that somewhere while adding that nr_vmemmap_pages is suitable.

Will come up with a way to 1) ensure that vmemmap pages is aligned and
2) disallow to use the feature in case it cannot be aligned for some reason.

Quoted 2 lines

Let's avoid using the "memblock" terminology. Just call it "get_nr_vmemmap_pages_cb" or maybe "detect_memmap_on_memory_cb"

Fine.

Quoted 13 lines

mm/memory_hotplug.c static int check_memblock_offlined_cb(struct memory_block *mem, void *arg) · line 1699

Quoted 6 lines
mm/memory_hotplug.c +6 -0
+	unsigned long *nr_vmemmap_pages = (unsigned long *)arg;
+	int ret = !mem->nr_vmemmap_pages;
+
+	if (!ret)
+		*nr_vmemmap_pages += mem->nr_vmemmap_pages;
+	return ret;

I think you can avoid "ret".

Also, we want to stop whenever we find "mem->nr_vmemmap_pages != 0", right? AFAIU, you would currently stop whenever you find mem->nr_vmemmap_pages == 0", which is wrong.

Quoted 6 lines
Quoted 3 lines

*nr_vmemmap_pages += mem->nr_vmemmap_pages;
/* Stop whenever we find any sign of MHP_MEMMAP_ON_MEMORY. */ return mem->nr_vmemmap_pages;

Yeah, true is that current logic is wrong, but we do not want to stop when mem->nr_vmemmap_pages == 0 either.

Yeah, as the comment said "Stop whenever we find any sign of MHP_MEMMAP_ON_MEMORY", whcih is the case with "mem->nr_vmemmap_pages", no?

Quoted 5 lines

Since we want to avoid people to remove with a different granularity than it was added, we need to check the whole memory range that is going to be hot-removed and check for mem->nr_vmemmap_pages to make sure that the range does not have any self-hosted vmemmap pages.

Quoted 8 lines

E.g:

[memory_block not using vmemmap] memblock#0
[memory block using vmemmap] memblock#1

So, if someone tries to remove memblock#0 and memblock#1 and once, we need to check whether any of the memblocks used self-hosted vmemmap, and if so, scream out loud if we are trying to remove site != memblock_size.

That's what my suggestion does, no?

Signature · 3 lines
Thanks,

David / dhildenb
author maintainer
#15 lore raw

Re: [PATCH 2/5] mm,memory_hotplug: Allocate memmap from the added memory range

Quoted 4 lines · David Hildenbrand

I'm confused.

1. Assume we hotplug memory, online it to ZONE_MOVABLE. The vmemmap gets allocated from altmap space.

The vmemmap could have never been allocated from altmap in case hpage vmemmap feature is enabled.

Have a look at [1].
If is_hugetlb_free_vmemmap_enabled(), vmemmap_populate() ends up calling vmemmap_populate_basepages().

And since no memory was consumed from altmap, and hence altmap_alloc_block_buf() was never called, vmem_altmap->alloc will be 0, and memory_block->nr_vmemmap_pages will be 0 as well.

But on a second though, true is that we will get in trouble if hpage vmemmap feature ever gets to work with vmemmap_populate_hugepages.
I will queue that to look in a new future.

[1] https://patchwork.kernel.org/project/linux-mm/patch/[email protected]/

Signature · 2 lines
Oscar Salvador
SUSE L3
#16 lore raw

Re: [PATCH 2/5] mm,memory_hotplug: Allocate memmap from the added memory range

Quoted 6 lines · Oscar Salvador
Quoted 4 lines · David Hildenbrand

I'm confused.

1. Assume we hotplug memory, online it to ZONE_MOVABLE. The vmemmap gets allocated from altmap space.

Quoted 6 lines

The vmemmap could have never been allocated from altmap in case hpage vmemmap feature is enabled.

Have a look at [1].
If is_hugetlb_free_vmemmap_enabled(), vmemmap_populate() ends up calling vmemmap_populate_basepages().

Oh, it calls "vmemmap_populate_basepages(start, end, node, NULL);"

the "NULL" part is the important bit.

Quoted 8 lines

And since no memory was consumed from altmap, and hence altmap_alloc_block_buf() was never called, vmem_altmap->alloc will be 0, and memory_block->nr_vmemmap_pages will be 0 as well.

But on a second though, true is that we will get in trouble if hpage vmemmap feature ever gets to work with vmemmap_populate_hugepages.
I will queue that to look in a new future.

This seriously needs comments and documentation. The problem is where to document as long as one of both series is not merged yet :)

At least in the cover letter, because this is not obvious how both things will play along.

Thanks!

Signature · 3 lines
Thanks,

David / dhildenb
author maintainer
#17 lore raw

PATCH 3/5 acpi,memhotplug: Enable MHP_MEMMAP_ON_MEMORY when supported

In reply to Oscar Salvador

Let the caller check whether it can pass MHP_MEMMAP_ON_MEMORY by checking mhp_supports_memmap_on_memory(). MHP_MEMMAP_ON_MEMORY can only be set in case ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE is enabled, the architecture supports altmap, and the range to be added spans a single memory block.

---
 drivers/acpi/acpi_memhotplug.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/acpi/acpi_memhotplug.c +4 -1
index b02fd51e5589..8cc195c4c861 100644
@@ -171,6 +171,7 @@ static int acpi_memory_enable_device(struct acpi_memory_device *mem_device)
 	acpi_handle handle = mem_device->device->handle;
 	int result, num_enabled = 0;
 	struct acpi_memory_info *info;
+	mhp_t mhp_flags = MHP_NONE;
 	int node;
 
 	node = acpi_get_node(handle);
@@ -194,8 +195,10 @@ static int acpi_memory_enable_device(struct acpi_memory_device *mem_device)
 		if (node < 0)
 			node = memory_add_physaddr_to_nid(info->start_addr);
 
+		if (mhp_supports_memmap_on_memory(info->length))
+			mhp_flags |= MHP_MEMMAP_ON_MEMORY;
 		result = __add_memory(node, info->start_addr, info->length,
-				      MHP_NONE);
+				      mhp_flags);
 
 		/*
 		 * If the memory block has been used by the kernel, add_memory()
Signature · 1 line
2.26.2
Reviewed-by on v1 3/5
#18 lore raw

Re: [PATCH 3/5] acpi,memhotplug: Enable MHP_MEMMAP_ON_MEMORY when supported

Quoted 26 lines · Oscar Salvador

Let the caller check whether it can pass MHP_MEMMAP_ON_MEMORY by checking mhp_supports_memmap_on_memory(). MHP_MEMMAP_ON_MEMORY can only be set in case ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE is enabled, the architecture supports altmap, and the range to be added spans a single memory block.

---
 drivers/acpi/acpi_memhotplug.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/acpi/acpi_memhotplug.c +1 -0
index b02fd51e5589..8cc195c4c861 100644
@@ -171,6 +171,7 @@ static int acpi_memory_enable_device(struct acpi_memory_device *mem_device)
 	acpi_handle handle = mem_device->device->handle;
 	int result, num_enabled = 0;
 	struct acpi_memory_info *info;
+	mhp_t mhp_flags = MHP_NONE;
 	int node;
 
 	node = acpi_get_node(handle);
@@ -194,8 +195,10 @@ static int acpi_memory_enable_device(struct acpi_memory_device *mem_device)
 		if (node < 0)

drivers/acpi/acpi_memhotplug.c static int acpi_memory_enable_device(struct acpi_memory_device *mem_device) · line 204

Quoted 10 lines
drivers/acpi/acpi_memhotplug.c +3 -1
 			node = memory_add_physaddr_to_nid(info->start_addr);
 
+		if (mhp_supports_memmap_on_memory(info->length))
+			mhp_flags |= MHP_MEMMAP_ON_MEMORY;
 		result = __add_memory(node, info->start_addr, info->length,
-				      MHP_NONE);
+				      mhp_flags);
 
 		/*
 		 * If the memory block has been used by the kernel, add_memory()
Signature · 3 lines
Thanks,

David / dhildenb
author maintainer
#19 lore raw

PATCH 4/5 powerpc/memhotplug: Enable MHP_MEMMAP_ON_MEMORY when supported

In reply to Oscar Salvador

Let the caller check whether it can pass MHP_MEMMAP_ON_MEMORY by checking mhp_supports_memmap_on_memory(). MHP_MEMMAP_ON_MEMORY can only be set in case ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE is enabled, the architecture supports altmap, and the range to be added spans a single memory block.

---
 arch/powerpc/platforms/pseries/hotplug-memory.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
arch/powerpc/platforms/pseries/hotplug-memory.c +4 -1
index 7efe6ec5d14a..a7f68e282ec1 100644
@@ -610,6 +610,7 @@ static int dlpar_memory_remove_by_ic(u32 lmbs_to_remove, u32 drc_index)
 
 static int dlpar_add_lmb(struct drmem_lmb *lmb)
 {
+	mhp_t mhp_flags = MHP_NONE;
 	unsigned long block_sz;
 	int nid, rc;
 
@@ -629,8 +630,10 @@ static int dlpar_add_lmb(struct drmem_lmb *lmb)
 	if (nid < 0 || !node_possible(nid))
 		nid = first_online_node;
 
+	if (mhp_supports_memmap_on_memory(block_sz))
+		mhp_flags |= MHP_MEMMAP_ON_MEMORY;
 	/* Add the memory */
-	rc = __add_memory(nid, lmb->base_addr, block_sz, MHP_NONE);
+	rc = __add_memory(nid, lmb->base_addr, block_sz, mhp_flags);
 	if (rc) {
 		invalidate_lmb_associativity_index(lmb);
 		return rc;
Signature · 1 line
2.26.2
#20 lore raw

Re: [PATCH 4/5] powerpc/memhotplug: Enable MHP_MEMMAP_ON_MEMORY when supported

Quoted 25 lines · Oscar Salvador

Let the caller check whether it can pass MHP_MEMMAP_ON_MEMORY by checking mhp_supports_memmap_on_memory(). MHP_MEMMAP_ON_MEMORY can only be set in case ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE is enabled, the architecture supports altmap, and the range to be added spans a single memory block.

---
 arch/powerpc/platforms/pseries/hotplug-memory.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
arch/powerpc/platforms/pseries/hotplug-memory.c +1 -0
index 7efe6ec5d14a..a7f68e282ec1 100644
@@ -610,6 +610,7 @@ static int dlpar_memory_remove_by_ic(u32 lmbs_to_remove, u32 drc_index)
 
 static int dlpar_add_lmb(struct drmem_lmb *lmb)
 {
+	mhp_t mhp_flags = MHP_NONE;
 	unsigned long block_sz;
 	int nid, rc;
 
@@ -629,8 +630,10 @@ static int dlpar_add_lmb(struct drmem_lmb *lmb)
 	if (nid < 0 || !node_possible(nid))

arch/powerpc/platforms/pseries/hotplug-memory.c static int dlpar_add_lmb(struct drmem_lmb *lmb) · line 639

Quoted 10 lines
arch/powerpc/platforms/pseries/hotplug-memory.c +3 -1
 		nid = first_online_node;
 
+	if (mhp_supports_memmap_on_memory(block_sz))
+		mhp_flags |= MHP_MEMMAP_ON_MEMORY;
 	/* Add the memory */
-	rc = __add_memory(nid, lmb->base_addr, block_sz, MHP_NONE);
+	rc = __add_memory(nid, lmb->base_addr, block_sz, mhp_flags);
 	if (rc) {
 		invalidate_lmb_associativity_index(lmb);
 		return rc;

With 16MB LMBs it's quite wasteful - you won't even have a huge page fitting the the remaining part.

I do wonder if we want this on powerpc only with a bigger LMB/memory block size (e.g., 256 MB, which is AFAIK the maximum usually found).

Signature · 3 lines
Thanks,

David / dhildenb
author maintainer
#21 lore raw

Re: [PATCH 4/5] powerpc/memhotplug: Enable MHP_MEMMAP_ON_MEMORY when supported

Quoted 7 lines · David Hildenbrand

On 17.12.20 14:07, Oscar Salvador wrote:

With 16MB LMBs it's quite wasteful - you won't even have a huge page fitting the the remaining part.

I do wonder if we want this on powerpc only with a bigger LMB/memory block size (e.g., 256 MB, which is AFAIK the maximum usually found).

Yeah, powerpc is trickier than the others.
It is more difficult to take advante of this in this platfowm as depending on the page's size, a larger memory are might have been mapped before, and so, further hot-add operations will be a noop wrt. populating sections, until the range to be added falls in non-mapped area.

Nevertheless, I will make this wired to larger memory blocks as you said.

Signature · 2 lines
Oscar Salvador
SUSE L3
author maintainer
#22 lore raw

PATCH 5/5 mm,memory_hotplug: Add kernel boot option to enable memmap_on_memory

In reply to Oscar Salvador

Self stored memmap leads to a sparse memory situation which is unsuitable for workloads that requires large contiguous memory chunks, so make this an opt-in which needs to be explicitly enabled.

To control this, let memory_hotplug have its own memory space, as suggested by David, so we can add memory_hotplug.memmap_on_memory parameter.

---
 .../admin-guide/kernel-parameters.txt         | 14 ++++++++++++
 mm/Makefile                                   |  5 ++++-
 mm/memory_hotplug.c                           | 22 ++++++++++++++++++-
 3 files changed, 39 insertions(+), 2 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +14 -0
index c722ec19cd00..8ff3d7c87165 100644
@@ -2776,6 +2776,20 @@
 			seconds.  Use this parameter to check at some
 			other rate.  0 disables periodic checking.
 
+	memory_hotplug.memmap_on_memory
+			[KNL,X86,ARM,PPC] Boolean flag to enable this feature.
+			Format: {on | off (default)}
+			When enabled, memory to build the pages tables for the
+			memmap array describing the hot-added range will be taken
+			from the range itself, so the memmap page tables will be
+			self-hosted.
+			Since only single memory device ranges are supported at
+			the moment, this option is disabled by default because
+			it might have an impact on workloads that needs large
+			contiguous memory chunks.
+			The state of the flag can be read in
+			/sys/module/memory_hotplug/parameters/memmap_on_memory.
+
 	memtest=	[KNL,X86,ARM,PPC] Enable memtest
 			Format: <integer>
 			default : 0 <disable>
mm/Makefile +4 -1
index 6b581f8337e8..0425cec6cb17 100644
@@ -58,9 +58,13 @@ obj-y			:= filemap.o mempool.o oom_kill.o fadvise.o \
 page-alloc-y := page_alloc.o
 page-alloc-$(CONFIG_SHUFFLE_PAGE_ALLOCATOR) += shuffle.o
 
+# Give 'memory_hotplug' its own module-parameter namespace
+memory-hotplug-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o
+
 obj-y += page-alloc.o
 obj-y += init-mm.o
 obj-y += memblock.o
+obj-y += $(memory-hotplug-y)
 
 ifdef CONFIG_MMU
 	obj-$(CONFIG_ADVISE_SYSCALLS)	+= madvise.o
@@ -83,7 +87,6 @@ obj-$(CONFIG_SLUB) += slub.o
 obj-$(CONFIG_KASAN)	+= kasan/
 obj-$(CONFIG_KFENCE) += kfence/
 obj-$(CONFIG_FAILSLAB) += failslab.o
-obj-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o
 obj-$(CONFIG_MEMTEST)		+= memtest.o
 obj-$(CONFIG_MIGRATION) += migrate.o
 obj-$(CONFIG_TRANSPARENT_HUGEPAGE) += huge_memory.o khugepaged.o
mm/memory_hotplug.c +21 -1
index 9371e7d3f583..f5f95f49b98a 100644
@@ -42,7 +42,27 @@
 #include "internal.h"
 #include "shuffle.h"
 
-static bool memmap_on_memory_enabled;
+/*
+ * memory_hotplug.memmap_on_memory parameter
+ */
+static bool memmap_on_memory_enabled __ro_after_init;
+
+static int memmap_on_memory_show(char *buffer, const struct kernel_param *kp)
+{
+	return sprintf(buffer, "%s\n",
+		       memmap_on_memory_enabled ? "on" : "off");
+}
+
+static __meminit int memmap_on_memory_store(const char *val,
+					    const struct kernel_param *kp)
+{
+	if (!IS_ENABLED(CONFIG_ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE))
+		return -EINVAL;
+
+	return param_set_bool(val, kp);
+}
+module_param_call(memmap_on_memory, memmap_on_memory_store,
+		  memmap_on_memory_show, &memmap_on_memory_enabled, 0400);
 
 /*
  * online_page_callback contains pointer to current page onlining function.
Signature · 1 line
2.26.2
author maintainer
#23 lore raw
In reply to Oscar Salvador
Quoted 12 lines · Oscar Salvador

I figured I would send a new version before going on vacation, so I can work on it when I am back.

Changes from RFCv3 to Patchv1:
- Addressed feedback from David
- Re-order patches

Changes from v2 -> v3:
- Re-order patches (Michal)
- Fold "mm,memory_hotplug: Introduce MHP_MEMMAP_ON_MEMORY" in patch#1
- Add kernel boot option to enable this feature (Michal)

Quoted 5 lines

Changes from v1 -> v2:
- Addressed feedback provided by David
- Add a arch_support_memmap_on_memory to be called from mhp_supports_memmap_on_memory, as atm, only ARM, powerpc and x86_64 have altmat support.

Now that Christmas holidays are gone, let us dig this up.

thanks

Signature · 2 lines
Oscar Salvador
SUSE L3