Re: [PATCH v2 1/2] driver/base: Optimize memory block registration to reduce boot time

[David Hildenbrand <david@redhat.com>]

On 28.04.25 19:03, Donet Tom wrote:
> During node device initialization, `memory blocks` are registered under
> each NUMA node. The `memory blocks` to be registered are identified using
> the node’s start and end PFNs, which are obtained from the node's pg_data
> 
> However, not all PFNs within this range necessarily belong to the same
> node—some may belong to other nodes. Additionally, due to the
> discontiguous nature of physical memory, certain sections within a
> `memory block` may be absent.> 
> As a result, `memory blocks` that fall between a node’s start and end
> PFNs may span across multiple nodes, and some sections within those blocks
> may be missing. `Memory blocks` have a fixed size, which is architecture
> dependent.
> 
> Due to these considerations, the memory block registration is currently
> performed as follows:
> 
> for_each_online_node(nid):
>      start_pfn = pgdat->node_start_pfn;
>      end_pfn = pgdat->node_start_pfn + node_spanned_pages;
>      for_each_memory_block_between(PFN_PHYS(start_pfn), PFN_PHYS(end_pfn))
>          mem_blk = memory_block_id(pfn_to_section_nr(pfn));
>          pfn_mb_start=section_nr_to_pfn(mem_blk->start_section_nr)
>          pfn_mb_end = pfn_start + memory_block_pfns - 1
>          for (pfn = pfn_mb_start; pfn < pfn_mb_end; pfn++):
>              if (get_nid_for_pfn(pfn) != nid):
>                  continue;
>              else
>                  do_register_memory_block_under_node(nid, mem_blk,
>                                                          MEMINIT_EARLY);
> 
> Here, we derive the start and end PFNs from the node's pg_data, then
> determine the memory blocks that may belong to the node. For each
> `memory block` in this range, we inspect all PFNs it contains and check
> their associated NUMA node ID. If a PFN within the block matches the
> current node, the memory block is registered under that node.
> 
> If CONFIG_DEFERRED_STRUCT_PAGE_INIT is enabled, get_nid_for_pfn() performs
> a binary search in the `memblock regions` to determine the NUMA node ID
> for a given PFN. If it is not enabled, the node ID is retrieved directly
> from the struct page.
> 
> On large systems, this process can become time-consuming, especially since
> we iterate over each `memory block` and all PFNs within it until a match is
> found. When CONFIG_DEFERRED_STRUCT_PAGE_INIT is enabled, the additional
> overhead of the binary search increases the execution time significantly,
> potentially leading to soft lockups during boot.
> 
> In this patch, we iterate over `memblock region` to identify the
> `memory blocks` that belong to the current NUMA node. `memblock regions`
> are contiguous memory ranges, each associated with a single NUMA node, and
> they do not span across multiple nodes.
> 
> for_each_online_node(nid):
>    for_each_memory_region(r): // r => region
>      if (r->nid != nid):
>        continue;
>      else
>        for_each_memory_block_between(r->base, r->base + r->size - 1):
>          do_register_memory_block_under_node(nid, mem_blk, MEMINIT_EARLY);
> 
> We iterate over all `memblock regions` and identify those that belong to
> the current NUMA node. For each `memblock region` associated with the
> current node, we calculate the start and end `memory blocks` based on the
> region's start and end PFNs. We then register all `memory blocks` within
> that range under the current node.

Yes, makes sense.

> 
> Test Results on My system with 32TB RAM
> =======================================
> 1. Boot time with CONFIG_DEFERRED_STRUCT_PAGE_INIT enabled.
> 
> Without this patch
> ------------------
> Startup finished in 1min 16.528s (kernel)
> 
> With this patch
> ---------------
> Startup finished in 17.236s (kernel) - 78% Improvement

Wow!

> 
> 2. Boot time with CONFIG_DEFERRED_STRUCT_PAGE_INIT disabled.
> 
> Without this patch
> ------------------
> Startup finished in 28.320s (kernel)
> 
> With this patch
> ---------------
> Startup finished in 15.621s (kernel) - 46% Improvement
> 

Also very nice!

> Signed-off-by: Donet Tom <donettom@linux.ibm.com>
> ---
> 
> v1->v2
> 
> Reworked the implementation according to suggestions from
> Mike Rapoport[1]
> 
> [1] - https://lore.kernel.org/all/Z_j2Gv9n4DOj6LSs@kernel.org/
> 
> v1 - https://lore.kernel.org/all/50142a29010463f436dc5c4feb540e5de3bb09df.1744175097.git.donettom@linux.ibm.com/
> ---
>   drivers/base/memory.c  |  4 ++--
>   drivers/base/node.c    | 39 +++++++++++++++++++++++++++++++++++++++
>   include/linux/memory.h |  2 ++
>   include/linux/node.h   | 11 +++++------
>   4 files changed, 48 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/base/memory.c b/drivers/base/memory.c
> index 19469e7f88c2..7f1d266ae593 100644
> --- a/drivers/base/memory.c
> +++ b/drivers/base/memory.c
> @@ -60,7 +60,7 @@ static inline unsigned long pfn_to_block_id(unsigned long pfn)
>   	return memory_block_id(pfn_to_section_nr(pfn));
>   }
>   
> -static inline unsigned long phys_to_block_id(unsigned long phys)
> +unsigned long phys_to_block_id(unsigned long phys)
>   {
>   	return pfn_to_block_id(PFN_DOWN(phys));
>   }
> @@ -632,7 +632,7 @@ int __weak arch_get_memory_phys_device(unsigned long start_pfn)
>    *
>    * Called under device_hotplug_lock.
>    */
> -static struct memory_block *find_memory_block_by_id(unsigned long block_id)
> +struct memory_block *find_memory_block_by_id(unsigned long block_id)
>   {
>   	struct memory_block *mem;
>   
> diff --git a/drivers/base/node.c b/drivers/base/node.c
> index cd13ef287011..4869333d366d 100644
> --- a/drivers/base/node.c
> +++ b/drivers/base/node.c
> @@ -20,6 +20,7 @@
>   #include <linux/pm_runtime.h>
>   #include <linux/swap.h>
>   #include <linux/slab.h>
> +#include <linux/memblock.h>
>   
>   static const struct bus_type node_subsys = {
>   	.name = "node",
> @@ -850,6 +851,44 @@ void unregister_memory_block_under_nodes(struct memory_block *mem_blk)
>   			  kobject_name(&node_devices[mem_blk->nid]->dev.kobj));
>   }
>   
> +/*
> + * register_memory_blocks_under_node_early : Register the memory
> + *		  blocks under the current node.
> + * @nid : Current node under registration
> + *
> + * This function iterates over all memblock regions and identifies the regions
> + * that belong to the current node. For each region which belongs to current
> + * node, it calculates the start and end memory blocks based on the region's
> + * start and end PFNs. It then registers all memory blocks within that range
> + * under the current node.
> + *
> + */
> +void register_memory_blocks_under_node_early(int nid)
> +{
> +	struct memblock_region *r;

You almost achieved a reverse x-mas tree :)

> +	unsigned long start_block_id;
> +	unsigned long end_block_id;
> +	struct memory_block *mem;
> +	unsigned long block_id;
> +
> +	for_each_mem_region(r) {
> +		if (r->nid == nid) {

To reduce indentation

if (r->nid != nid)
	continue;

> +			start_block_id = phys_to_block_id(r->base);
> +			end_block_id = phys_to_block_id(r->base + r->size - 1);

Probably you could make them const in the for loop

	const unsigned long start_block_id = phys_to_block_id(r->base);
	const unsigned long end_block_id = phys_to_block_id(r->base + r->size - 1);

Okay, so end is inclusive as well.

> +
> +			for (block_id = start_block_id; block_id <= end_block_id; block_id++) {
> +				mem = find_memory_block_by_id(block_id);
> +				if (!mem)
> +					continue;
> +
> +				do_register_memory_block_under_node(nid, mem, MEMINIT_EARLY);

-- 
Cheers,

David / dhildenb