Page Pool API

The page_pool allocator is optimized for recycling page or page fragment used by skb packet and xdp frame.

Basic use involves replacing any alloc_pages() calls with page_pool_alloc(), which allocate memory with or without page splitting depending on the requested memory size.

If the driver knows that it always requires full pages or its allocations are always smaller than half a page, it can use one of the more specific API calls:

1. page_pool_alloc_pages(): allocate memory without page splitting when driver knows that the memory it need is always bigger than half of the page allocated from page pool. There is no cache line dirtying for 'struct page' when a page is recycled back to the page pool.

2. page_pool_alloc_frag(): allocate memory with page splitting when driver knows that the memory it need is always smaller than or equal to half of the page allocated from page pool. Page splitting enables memory saving and thus avoids TLB/cache miss for data access, but there also is some cost to implement page splitting, mainly some cache line dirtying/bouncing for 'struct page' and atomic operation for page->pp_ref_count.

The API keeps track of in-flight pages, in order to let API users know when it is safe to free a page_pool object, the API users must call page_pool_put_page() or page_pool_free_va() to free the page_pool object, or attach the page_pool object to a page_pool-aware object like skbs marked with skb_mark_for_recycle().

page_pool_put_page() may be called multiple times on the same page if a page is split into multiple fragments. For the last fragment, it will either recycle the page, or in case of page->_refcount > 1, it will release the DMA mapping and in-flight state accounting.

dma_sync_single_range_for_device() is only called for the last fragment when page_pool is created with PP_FLAG_DMA_SYNC_DEV flag, so it depends on the last freed fragment to do the sync_for_device operation for all fragments in the same page when a page is split. The API user must setup pool->p.max_len and pool->p.offset correctly and ensure that page_pool_put_page() is called with dma_sync_size being -1 for fragment API.

Architecture overview

+------------------+
|       Driver     |
+------------------+
        ^
        |
        |
        |
        v
+--------------------------------------------+
|                request memory              |
+--------------------------------------------+
    ^                                  ^
    |                                  |
    | Pool empty                       | Pool has entries
    |                                  |
    v                                  v
+-----------------------+     +------------------------+
| alloc (and map) pages |     |  get page from cache   |
+-----------------------+     +------------------------+
                                ^                    ^
                                |                    |
                                | cache available    | No entries, refill
                                |                    | from ptr-ring
                                |                    |
                                v                    v
                      +-----------------+     +------------------+
                      |   Fast cache    |     |  ptr-ring cache  |
                      +-----------------+     +------------------+

Monitoring

Information about page pools on the system can be accessed via the netdev genetlink family (see Documentation/netlink/specs/netdev.yaml).

API interface

The number of pools created must match the number of hardware queues unless hardware restrictions make that impossible. This would otherwise beat the purpose of page pool, which is allocate pages fast from cache without locking. This lockless guarantee naturally comes from running under a NAPI softirq. The protection doesn't strictly have to be NAPI, any guarantee that allocating a page will cause no race conditions is enough.

struct page_pool *page_pool_create(const struct page_pool_params *params)

create a page pool.

Parameters

const struct page_pool_params *params

parameters, see struct page_pool_params

struct page_pool_params

page pool parameters

Definition:

struct page_pool_params {
    unsigned int    flags;
    unsigned int    order;
    unsigned int    pool_size;
    int nid;
    struct device   *dev;
    struct napi_struct *napi;
    enum dma_data_direction dma_dir;
    unsigned int    max_len;
    unsigned int    offset;
    STRUCT_GROUP( struct net_device *netdev;
};

Members

flags

PP_FLAG_DMA_MAP, PP_FLAG_DMA_SYNC_DEV

order

2^order pages on allocation

pool_size

size of the ptr_ring

nid

NUMA node id to allocate from pages from

dev

device, for DMA pre-mapping purposes

napi

NAPI which is the sole consumer of pages, otherwise NULL

dma_dir

DMA mapping direction

max_len

max DMA sync memory size for PP_FLAG_DMA_SYNC_DEV

offset

DMA sync address offset for PP_FLAG_DMA_SYNC_DEV

netdev

netdev this pool will serve (leave as NULL if none or multiple)

struct page *page_pool_dev_alloc_pages(struct page_pool *pool)

allocate a page.

Parameters

struct page_pool *pool

pool from which to allocate

Description

Get a page from the page allocator or page_pool caches.

struct page *page_pool_dev_alloc_frag(struct page_pool *pool, unsigned int *offset, unsigned int size)

allocate a page fragment.

Parameters

struct page_pool *pool

pool from which to allocate

unsigned int *offset

offset to the allocated page

unsigned int size

requested size

Description

Get a page fragment from the page allocator or page_pool caches.

Return

Return allocated page fragment, otherwise return NULL.

struct page *page_pool_dev_alloc(struct page_pool *pool, unsigned int *offset, unsigned int *size)

allocate a page or a page fragment.

Parameters

struct page_pool *pool

pool from which to allocate

unsigned int *offset

offset to the allocated page

unsigned int *size

in as the requested size, out as the allocated size

Description

Get a page or a page fragment from the page allocator or page_pool caches depending on the requested size in order to allocate memory with least memory utilization and performance penalty.

Return

Return allocated page or page fragment, otherwise return NULL.

void *page_pool_dev_alloc_va(struct page_pool *pool, unsigned int *size)

allocate a page or a page fragment and return its va.

Parameters

struct page_pool *pool

pool from which to allocate

unsigned int *size

in as the requested size, out as the allocated size

Description

This is just a thin wrapper around the page_pool_alloc() API, and it returns va of the allocated page or page fragment.

Return

Return the va for the allocated page or page fragment, otherwise return NULL.

enum dma_data_direction page_pool_get_dma_dir(struct page_pool *pool)

Retrieve the stored DMA direction.

Parameters

struct page_pool *pool

pool from which page was allocated

Description

Get the stored dma direction. A driver might decide to store this locally and avoid the extra cache line from page_pool to determine the direction.

void page_pool_put_page(struct page_pool *pool, struct page *page, unsigned int dma_sync_size, bool allow_direct)

release a reference to a page pool page

Parameters

struct page_pool *pool

pool from which page was allocated

struct page *page

page to release a reference on

unsigned int dma_sync_size

how much of the page may have been touched by the device

bool allow_direct

released by the consumer, allow lockless caching

Description

The outcome of this depends on the page refcnt. If the driver bumps the refcnt > 1 this will unmap the page. If the page refcnt is 1 the allocator owns the page and will try to recycle it in one of the pool caches. If PP_FLAG_DMA_SYNC_DEV is set, the page will be synced for_device using dma_sync_single_range_for_device().

void page_pool_put_full_page(struct page_pool *pool, struct page *page, bool allow_direct)

release a reference on a page pool page

Parameters

struct page_pool *pool

pool from which page was allocated

struct page *page

page to release a reference on

bool allow_direct

released by the consumer, allow lockless caching

Description

Similar to page_pool_put_page(), but will DMA sync the entire memory area as configured in page_pool_params.max_len.

void page_pool_recycle_direct(struct page_pool *pool, struct page *page)

release a reference on a page pool page

Parameters

struct page_pool *pool

pool from which page was allocated

struct page *page

page to release a reference on

Description

Similar to page_pool_put_full_page() but caller must guarantee safe context (e.g NAPI), since it will recycle the page directly into the pool fast cache.

void page_pool_free_va(struct page_pool *pool, void *va, bool allow_direct)

free a va into the page_pool

Parameters

struct page_pool *pool

pool from which va was allocated

void *va

va to be freed

bool allow_direct

freed by the consumer, allow lockless caching

Description

Free a va allocated from page_pool_allo_va().

dma_addr_t page_pool_get_dma_addr(struct page *page)

Retrieve the stored DMA address.

Parameters

struct page *page

page allocated from a page pool

Description

Fetch the DMA address of the page. The page pool to which the page belongs must had been created with PP_FLAG_DMA_MAP.

bool page_pool_get_stats(const struct page_pool *pool, struct page_pool_stats *stats)

fetch page pool stats

Parameters

const struct page_pool *pool

pool from which page was allocated

struct page_pool_stats *stats

struct page_pool_stats to fill in

Description

Retrieve statistics about the page_pool. This API is only available if the kernel has been configured with CONFIG_PAGE_POOL_STATS=y. A pointer to a caller allocated struct page_pool_stats structure is passed to this API which is filled in. The caller can then report those stats to the user (perhaps via ethtool, debugfs, etc.).

void page_pool_put_page_bulk(struct page_pool *pool, void **data, int count)

release references on multiple pages

Parameters

struct page_pool *pool

pool from which pages were allocated

void **data

array holding page pointers

int count

number of pages in data

Description

Tries to refill a number of pages into the ptr_ring cache holding ptr_ring producer lock. If the ptr_ring is full, page_pool_put_page_bulk() will release leftover pages to the page allocator. page_pool_put_page_bulk() is suitable to be run inside the driver NAPI tx completion loop for the XDP_REDIRECT use case.

Please note the caller must not use data area after running page_pool_put_page_bulk(), as this function overwrites it.

DMA sync

Driver is always responsible for syncing the pages for the CPU. Drivers may choose to take care of syncing for the device as well or set the PP_FLAG_DMA_SYNC_DEV flag to request that pages allocated from the page pool are already synced for the device.

If PP_FLAG_DMA_SYNC_DEV is set, the driver must inform the core what portion of the buffer has to be synced. This allows the core to avoid syncing the entire page when the drivers knows that the device only accessed a portion of the page.

Most drivers will reserve headroom in front of the frame. This part of the buffer is not touched by the device, so to avoid syncing it drivers can set the offset field in struct page_pool_params appropriately.

For pages recycled on the XDP xmit and skb paths the page pool will use the max_len member of struct page_pool_params to decide how much of the page needs to be synced (starting at offset). When directly freeing pages in the driver (page_pool_put_page()) the dma_sync_size argument specifies how much of the buffer needs to be synced.

If in doubt set offset to 0, max_len to PAGE_SIZE and pass -1 as dma_sync_size. That combination of arguments is always correct.

Note that the syncing parameters are for the entire page. This is important to remember when using fragments (PP_FLAG_PAGE_FRAG), where allocated buffers may be smaller than a full page. Unless the driver author really understands page pool internals it's recommended to always use offset = 0, max_len = PAGE_SIZE with fragmented page pools.

Stats API and structures

If the kernel is configured with CONFIG_PAGE_POOL_STATS=y, the API page_pool_get_stats() and structures described below are available. It takes a pointer to a struct page_pool and a pointer to a struct page_pool_stats allocated by the caller.

Older drivers expose page pool statistics via ethtool or debugfs. The same statistics are accessible via the netlink netdev family in a driver-independent fashion.

struct page_pool_alloc_stats

allocation statistics

Definition:

struct page_pool_alloc_stats {
    u64 fast;
    u64 slow;
    u64 slow_high_order;
    u64 empty;
    u64 refill;
    u64 waive;
};

Members

fast

successful fast path allocations

slow

slow path order-0 allocations

slow_high_order

slow path high order allocations

empty

ptr ring is empty, so a slow path allocation was forced

refill

an allocation which triggered a refill of the cache

waive

pages obtained from the ptr ring that cannot be added to the cache due to a NUMA mismatch

struct page_pool_recycle_stats

recycling (freeing) statistics

Definition:

struct page_pool_recycle_stats {
    u64 cached;
    u64 cache_full;
    u64 ring;
    u64 ring_full;
    u64 released_refcnt;
};

Members

cached

recycling placed page in the page pool cache

cache_full

page pool cache was full

ring

page placed into the ptr ring

ring_full

page released from page pool because the ptr ring was full

released_refcnt

page released (and not recycled) because refcnt > 1

struct page_pool_stats

combined page pool use statistics

Definition:

struct page_pool_stats {
    struct page_pool_alloc_stats alloc_stats;
    struct page_pool_recycle_stats recycle_stats;
};

Members

alloc_stats

see struct page_pool_alloc_stats

recycle_stats

see struct page_pool_recycle_stats

Description

Wrapper struct for combining page pool stats with different storage requirements.

Coding examples

Registration

/* Page pool registration */
struct page_pool_params pp_params = { 0 };
struct xdp_rxq_info xdp_rxq;
int err;

pp_params.order = 0;
/* internal DMA mapping in page_pool */
pp_params.flags = PP_FLAG_DMA_MAP;
pp_params.pool_size = DESC_NUM;
pp_params.nid = NUMA_NO_NODE;
pp_params.dev = priv->dev;
pp_params.napi = napi; /* only if locking is tied to NAPI */
pp_params.dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
page_pool = page_pool_create(&pp_params);

err = xdp_rxq_info_reg(&xdp_rxq, ndev, 0);
if (err)
    goto err_out;

err = xdp_rxq_info_reg_mem_model(&xdp_rxq, MEM_TYPE_PAGE_POOL, page_pool);
if (err)
    goto err_out;

NAPI poller

/* NAPI Rx poller */
enum dma_data_direction dma_dir;

dma_dir = page_pool_get_dma_dir(dring->page_pool);
while (done < budget) {
    if (some error)
        page_pool_recycle_direct(page_pool, page);
    if (packet_is_xdp) {
        if XDP_DROP:
            page_pool_recycle_direct(page_pool, page);
    } else (packet_is_skb) {
        skb_mark_for_recycle(skb);
        new_page = page_pool_dev_alloc_pages(page_pool);
    }
}

Stats

#ifdef CONFIG_PAGE_POOL_STATS
/* retrieve stats */
struct page_pool_stats stats = { 0 };
if (page_pool_get_stats(page_pool, &stats)) {
        /* perhaps the driver reports statistics with ethool */
        ethtool_print_allocation_stats(&stats.alloc_stats);
        ethtool_print_recycle_stats(&stats.recycle_stats);
}
#endif

Driver unload

/* Driver unload */
page_pool_put_full_page(page_pool, page, false);
xdp_rxq_info_unreg(&xdp_rxq);