Re: [PATCH net-next] page_pool: disable direct recycling based on pool->cpuid on destroy

From: Lorenzo Bianconi
Date: Thu Feb 15 2024 - 07:16:30 EST


> Now that direct recycling is performed basing on pool->cpuid when set,
> memory leaks are possible:
>
> 1. A pool is destroyed.
> 2. Alloc cache is emptied (it's done only once).
> 3. pool->cpuid is still set.
> 4. napi_pp_put_page() does direct recycling basing on pool->cpuid.
> 5. Now alloc cache is not empty, but it won't ever be freed.
>
> In order to avoid that, rewrite pool->cpuid to -1 when unlinking NAPI to
> make sure no direct recycling will be possible after emptying the cache.
> This involves a bit of overhead as pool->cpuid now must be accessed
> via READ_ONCE() to avoid partial reads.
> Rename page_pool_unlink_napi() -> page_pool_disable_direct_recycling()
> to reflect what it actually does and unexport it.

Hi Alexander,

IIUC the reported issue, it requires the page_pool is destroyed (correct?),
but system page_pool (the only one with cpuid not set to -1) will never be
destroyed at runtime (or at we should avoid that). Am I missing something?

Rergards,
Lorenzo

>
> Fixes: 2b0cfa6e4956 ("net: add generic percpu page_pool allocator")
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@xxxxxxxxx>
> ---
> include/net/page_pool/types.h | 5 -----
> net/core/page_pool.c | 10 +++++++---
> net/core/skbuff.c | 2 +-
> 3 files changed, 8 insertions(+), 9 deletions(-)
>
> diff --git a/include/net/page_pool/types.h b/include/net/page_pool/types.h
> index 3828396ae60c..3590fbe6e3f1 100644
> --- a/include/net/page_pool/types.h
> +++ b/include/net/page_pool/types.h
> @@ -210,17 +210,12 @@ struct page_pool *page_pool_create_percpu(const struct page_pool_params *params,
> struct xdp_mem_info;
>
> #ifdef CONFIG_PAGE_POOL
> -void page_pool_unlink_napi(struct page_pool *pool);
> void page_pool_destroy(struct page_pool *pool);
> void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *),
> struct xdp_mem_info *mem);
> void page_pool_put_page_bulk(struct page_pool *pool, void **data,
> int count);
> #else
> -static inline void page_pool_unlink_napi(struct page_pool *pool)
> -{
> -}
> -
> static inline void page_pool_destroy(struct page_pool *pool)
> {
> }
> diff --git a/net/core/page_pool.c b/net/core/page_pool.c
> index 89c835fcf094..e8b9399d8e32 100644
> --- a/net/core/page_pool.c
> +++ b/net/core/page_pool.c
> @@ -949,8 +949,13 @@ void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *),
> pool->xdp_mem_id = mem->id;
> }
>
> -void page_pool_unlink_napi(struct page_pool *pool)
> +static void page_pool_disable_direct_recycling(struct page_pool *pool)
> {
> + /* Disable direct recycling based on pool->cpuid.
> + * Paired with READ_ONCE() in napi_pp_put_page().
> + */
> + WRITE_ONCE(pool->cpuid, -1);
> +
> if (!pool->p.napi)
> return;
>
> @@ -962,7 +967,6 @@ void page_pool_unlink_napi(struct page_pool *pool)
>
> WRITE_ONCE(pool->p.napi, NULL);
> }
> -EXPORT_SYMBOL(page_pool_unlink_napi);
>
> void page_pool_destroy(struct page_pool *pool)
> {
> @@ -972,7 +976,7 @@ void page_pool_destroy(struct page_pool *pool)
> if (!page_pool_put(pool))
> return;
>
> - page_pool_unlink_napi(pool);
> + page_pool_disable_direct_recycling(pool);
> page_pool_free_frag(pool);
>
> if (!page_pool_release(pool))
> diff --git a/net/core/skbuff.c b/net/core/skbuff.c
> index 0d9a489e6ae1..b41856585c24 100644
> --- a/net/core/skbuff.c
> +++ b/net/core/skbuff.c
> @@ -1018,7 +1018,7 @@ bool napi_pp_put_page(struct page *page, bool napi_safe)
> unsigned int cpuid = smp_processor_id();
>
> allow_direct = napi && READ_ONCE(napi->list_owner) == cpuid;
> - allow_direct |= (pp->cpuid == cpuid);
> + allow_direct |= READ_ONCE(pp->cpuid) == cpuid;
> }
>
> /* Driver set this to memory recycling info. Reset it on recycle.
> --
> 2.43.0
>

Attachment: signature.asc
Description: PGP signature