From patchwork Thu Sep 12 12:45:13 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Yunsheng Lin X-Patchwork-Id: 13802105 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 0ED90EEB59F for ; Thu, 12 Sep 2024 12:55:35 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender:List-Subscribe:List-Help :List-Post:List-Archive:List-Unsubscribe:List-Id:Content-Type: Content-Transfer-Encoding:MIME-Version:References:In-Reply-To:Message-ID:Date :Subject:CC:To:From:Reply-To:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Owner; bh=xiPUboos7UnPPyrhjvzDt9vrPjPpHrI8AhzIAyKodeU=; b=CZiYUiGbjIdJK9ZcLi5IepvWEI bwB1XKjCyzVqrkl5gcSu6vGGUnNXt/sZj91ItQF4DI/vj+fhHz6Z5IgHg8IlElOySRkCLVnkYbYaE 6gIxQnCiACZ6q508+qUpXl+mCKOMoJNU0cD4RM2YMqWYSRZV+ysbW5viJkMfyusMT1/HKTHV7ZIRb r1QYPWJEOKk+GZp9BNK0JoNCB1t8kvJ9Ir+9esoFI1MoPI+oqBSB4zShqJfPV9o8zLF7xKEbWRb2w tK1/7tN6rwghyveKLh/g9F3m0jzNeHDsDENPyI0pelC2olo05JUZ7wgjWd3cfOLQf8xK1SjRO2RCW P4W0nyIg==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.97.1 #2 (Red Hat Linux)) id 1sojM0-0000000D70J-28I2; Thu, 12 Sep 2024 12:55:20 +0000 Received: from szxga01-in.huawei.com ([45.249.212.187]) by bombadil.infradead.org with esmtps (Exim 4.97.1 #2 (Red Hat Linux)) id 1sojI6-0000000D5zT-3GDJ; Thu, 12 Sep 2024 12:51:21 +0000 Received: from mail.maildlp.com (unknown [172.19.163.48]) by szxga01-in.huawei.com (SkyGuard) with ESMTP id 4X4HNN25z0zyRkF; Thu, 12 Sep 2024 20:50:28 +0800 (CST) Received: from dggpemf200006.china.huawei.com (unknown [7.185.36.61]) by mail.maildlp.com (Postfix) with ESMTPS id 3112D180064; Thu, 12 Sep 2024 20:51:15 +0800 (CST) Received: from localhost.localdomain (10.90.30.45) by dggpemf200006.china.huawei.com (7.185.36.61) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.11; Thu, 12 Sep 2024 20:51:14 +0800 From: Yunsheng Lin To: , , CC: , , Yunsheng Lin , Robin Murphy , Alexander Duyck , IOMMU , Wei Fang , Shenwei Wang , Clark Wang , Eric Dumazet , Tony Nguyen , Przemek Kitszel , Alexander Lobakin , Alexei Starovoitov , Daniel Borkmann , Jesper Dangaard Brouer , John Fastabend , Saeed Mahameed , Leon Romanovsky , Tariq Toukan , Felix Fietkau , Lorenzo Bianconi , Ryder Lee , Shayne Chen , Sean Wang , Kalle Valo , Matthias Brugger , AngeloGioacchino Del Regno , Andrew Morton , Ilias Apalodimas , , , , , , , , , , Subject: [RFC 2/2] page_pool: fix IOMMU crash when driver has already unbound Date: Thu, 12 Sep 2024 20:45:13 +0800 Message-ID: <20240912124514.2329991-3-linyunsheng@huawei.com> X-Mailer: git-send-email 2.30.0 In-Reply-To: <20240912124514.2329991-1-linyunsheng@huawei.com> References: <20240912124514.2329991-1-linyunsheng@huawei.com> MIME-Version: 1.0 X-Originating-IP: [10.90.30.45] X-ClientProxiedBy: dggems702-chm.china.huawei.com (10.3.19.179) To dggpemf200006.china.huawei.com (7.185.36.61) X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20240912_055119_393077_F7F2AE83 X-CRM114-Status: GOOD ( 27.81 ) X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=archiver.kernel.org@lists.infradead.org Networking driver with page_pool support may hand over page still with dma mapping to network stack and try to reuse that page after network stack is done with it and passes it back to page_pool to avoid the penalty of dma mapping/unmapping. With all the caching in the network stack, some pages may be held in the network stack without returning to the page_pool soon enough, and with VF disable causing the driver unbound, the page_pool does not stop the driver from doing it's unbounding work, instead page_pool uses workqueue to check if there is some pages coming back from the network stack periodically, if there is any, it will do the dma unmmapping related cleanup work. As mentioned in [1], attempting DMA unmaps after the driver has already unbound may leak resources or at worst corrupt memory. Fundamentally, the page pool code cannot allow DMA mappings to outlive the driver they belong to. Currently it seems there are at least two cases that the page is not released fast enough causing dma unmmapping done after driver has already unbound: 1. ipv4 packet defragmentation timeout: this seems to cause delay up to 30 secs: 2. skb_defer_free_flush(): this may cause infinite delay if there is no triggering for net_rx_action(). In order not to do the dma unmmapping after driver has already unbound and stall the unloading of the networking driver, add the pool->items array to record all the pages including the ones which are handed over to network stack, so the page_pool can do the dma unmmapping for those pages when page_pool_destroy() is called. Note, the devmem patchset seems to make the bug harder to fix and to backport too, this patch does not consider fixing the case for devmem yet. 1. https://lore.kernel.org/lkml/8067f204-1380-4d37-8ffd-007fc6f26738@kernel.org/T/ Signed-off-by: Yunsheng Lin CC: Robin Murphy CC: Alexander Duyck CC: IOMMU --- drivers/net/ethernet/freescale/fec_main.c | 8 +- drivers/net/ethernet/intel/iavf/iavf_txrx.c | 6 +- drivers/net/ethernet/intel/idpf/idpf_txrx.c | 14 +- drivers/net/ethernet/intel/libeth/rx.c | 2 +- .../net/ethernet/mellanox/mlx5/core/en/xdp.c | 3 +- drivers/net/netdevsim/netdev.c | 6 +- drivers/net/wireless/mediatek/mt76/mt76.h | 2 +- include/linux/mm_types.h | 2 +- include/linux/skbuff.h | 1 + include/net/libeth/rx.h | 3 +- include/net/netmem.h | 2 +- include/net/page_pool/helpers.h | 11 ++ include/net/page_pool/types.h | 15 +- net/core/devmem.c | 2 +- net/core/page_pool.c | 159 +++++++++++++++--- net/core/page_pool_priv.h | 9 +- net/core/skbuff.c | 3 +- net/core/xdp.c | 3 +- 18 files changed, 202 insertions(+), 49 deletions(-) diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index acbb627d51bf..c00f8c460759 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -1009,7 +1009,8 @@ static void fec_enet_bd_init(struct net_device *dev) struct page *page = txq->tx_buf[i].buf_p; if (page) - page_pool_put_page(page->pp, page, 0, false); + page_pool_put_page(page_pool_to_pp(page), + page, 0, false); } txq->tx_buf[i].buf_p = NULL; @@ -1538,7 +1539,7 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id, int budget) xdp_return_frame_rx_napi(xdpf); } else { /* recycle pages of XDP_TX frames */ /* The dma_sync_size = 0 as XDP_TX has already synced DMA for_device */ - page_pool_put_page(page->pp, page, 0, true); + page_pool_put_page(page_pool_to_pp(page), page, 0, true); } txq->tx_buf[index].buf_p = NULL; @@ -3300,7 +3301,8 @@ static void fec_enet_free_buffers(struct net_device *ndev) } else { struct page *page = txq->tx_buf[i].buf_p; - page_pool_put_page(page->pp, page, 0, false); + page_pool_put_page(page_pool_to_pp(page), + page, 0, false); } txq->tx_buf[i].buf_p = NULL; diff --git a/drivers/net/ethernet/intel/iavf/iavf_txrx.c b/drivers/net/ethernet/intel/iavf/iavf_txrx.c index 26b424fd6718..658d8f9a6abb 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_txrx.c +++ b/drivers/net/ethernet/intel/iavf/iavf_txrx.c @@ -1050,7 +1050,8 @@ static void iavf_add_rx_frag(struct sk_buff *skb, const struct libeth_fqe *rx_buffer, unsigned int size) { - u32 hr = rx_buffer->page->pp->p.offset; + struct page_pool *pool = page_pool_to_pp(rx_buffer->page); + u32 hr = pool->p.offset; skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buffer->page, rx_buffer->offset + hr, size, rx_buffer->truesize); @@ -1067,7 +1068,8 @@ static void iavf_add_rx_frag(struct sk_buff *skb, static struct sk_buff *iavf_build_skb(const struct libeth_fqe *rx_buffer, unsigned int size) { - u32 hr = rx_buffer->page->pp->p.offset; + struct page_pool *pool = page_pool_to_pp(rx_buffer->page); + u32 hr = pool->p.offset; struct sk_buff *skb; void *va; diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c index d4e6f0e10487..1a4d83b8fda7 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c @@ -385,7 +385,8 @@ static void idpf_rx_page_rel(struct libeth_fqe *rx_buf) if (unlikely(!rx_buf->page)) return; - page_pool_put_full_page(rx_buf->page->pp, rx_buf->page, false); + page_pool_put_full_page(page_pool_to_pp(rx_buf->page), rx_buf->page, + false); rx_buf->page = NULL; rx_buf->offset = 0; @@ -3097,7 +3098,8 @@ idpf_rx_process_skb_fields(struct idpf_rx_queue *rxq, struct sk_buff *skb, void idpf_rx_add_frag(struct idpf_rx_buf *rx_buf, struct sk_buff *skb, unsigned int size) { - u32 hr = rx_buf->page->pp->p.offset; + struct page_pool *pool = page_pool_to_pp(rx_buffer->page); + u32 hr = pool->p.offset; skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buf->page, rx_buf->offset + hr, size, rx_buf->truesize); @@ -3129,8 +3131,10 @@ static u32 idpf_rx_hsplit_wa(const struct libeth_fqe *hdr, if (!libeth_rx_sync_for_cpu(buf, copy)) return 0; - dst = page_address(hdr->page) + hdr->offset + hdr->page->pp->p.offset; - src = page_address(buf->page) + buf->offset + buf->page->pp->p.offset; + dst = page_address(hdr->page) + hdr->offset + + page_pool_to_pp(hdr->page)->p.offset; + src = page_address(buf->page) + buf->offset + + page_pool_to_pp(buf->page)->p.offset; memcpy(dst, src, LARGEST_ALIGN(copy)); buf->offset += copy; @@ -3148,7 +3152,7 @@ static u32 idpf_rx_hsplit_wa(const struct libeth_fqe *hdr, */ struct sk_buff *idpf_rx_build_skb(const struct libeth_fqe *buf, u32 size) { - u32 hr = buf->page->pp->p.offset; + u32 hr = page_pool_to_pp(buf->page)->p.offset; struct sk_buff *skb; void *va; diff --git a/drivers/net/ethernet/intel/libeth/rx.c b/drivers/net/ethernet/intel/libeth/rx.c index f20926669318..385afca0e61d 100644 --- a/drivers/net/ethernet/intel/libeth/rx.c +++ b/drivers/net/ethernet/intel/libeth/rx.c @@ -207,7 +207,7 @@ EXPORT_SYMBOL_NS_GPL(libeth_rx_fq_destroy, LIBETH); */ void libeth_rx_recycle_slow(struct page *page) { - page_pool_recycle_direct(page->pp, page); + page_pool_recycle_direct(page_pool_to_pp(page), page); } EXPORT_SYMBOL_NS_GPL(libeth_rx_recycle_slow, LIBETH); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index 4610621a340e..83511a45a6dc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -716,7 +716,8 @@ static void mlx5e_free_xdpsq_desc(struct mlx5e_xdpsq *sq, /* No need to check ((page->pp_magic & ~0x3UL) == PP_SIGNATURE) * as we know this is a page_pool page. */ - page_pool_recycle_direct(page->pp, page); + page_pool_recycle_direct(page_pool_to_pp(page), + page); } while (++n < num); break; diff --git a/drivers/net/netdevsim/netdev.c b/drivers/net/netdevsim/netdev.c index 017a6102be0a..9bfa593cd5dd 100644 --- a/drivers/net/netdevsim/netdev.c +++ b/drivers/net/netdevsim/netdev.c @@ -593,7 +593,8 @@ nsim_pp_hold_write(struct file *file, const char __user *data, if (!ns->page) ret = -ENOMEM; } else { - page_pool_put_full_page(ns->page->pp, ns->page, false); + page_pool_put_full_page(page_pool_to_pp(ns->page), ns->page, + false); ns->page = NULL; } rtnl_unlock(); @@ -788,7 +789,8 @@ void nsim_destroy(struct netdevsim *ns) /* Put this intentionally late to exercise the orphaning path */ if (ns->page) { - page_pool_put_full_page(ns->page->pp, ns->page, false); + page_pool_put_full_page(page_pool_to_pp(ns->page), ns->page, + false); ns->page = NULL; } diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h index 0b75a45ad2e8..94a277290909 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76.h +++ b/drivers/net/wireless/mediatek/mt76/mt76.h @@ -1688,7 +1688,7 @@ static inline void mt76_put_page_pool_buf(void *buf, bool allow_direct) { struct page *page = virt_to_head_page(buf); - page_pool_put_full_page(page->pp, page, allow_direct); + page_pool_put_full_page(page_pool_to_pp(page), page, allow_direct); } static inline void * diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 485424979254..410187133d27 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -120,7 +120,7 @@ struct page { * page_pool allocated pages. */ unsigned long pp_magic; - struct page_pool *pp; + struct page_pool_item *pp_item; unsigned long _pp_mapping_pad; unsigned long dma_addr; atomic_long_t pp_ref_count; diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 39f1d16f3628..64d1ecb7a7fc 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -38,6 +38,7 @@ #include #include #include +#include /** * DOC: skb checksums diff --git a/include/net/libeth/rx.h b/include/net/libeth/rx.h index 43574bd6612f..beee7ddd77a5 100644 --- a/include/net/libeth/rx.h +++ b/include/net/libeth/rx.h @@ -137,7 +137,8 @@ static inline bool libeth_rx_sync_for_cpu(const struct libeth_fqe *fqe, return false; } - page_pool_dma_sync_for_cpu(page->pp, page, fqe->offset, len); + page_pool_dma_sync_for_cpu(page_pool_to_pp(page), page, fqe->offset, + len); return true; } diff --git a/include/net/netmem.h b/include/net/netmem.h index 8a6e20be4b9d..9bd2db627dbb 100644 --- a/include/net/netmem.h +++ b/include/net/netmem.h @@ -49,7 +49,7 @@ struct net_iov { static_assert(offsetof(struct page, pg) == \ offsetof(struct net_iov, iov)) NET_IOV_ASSERT_OFFSET(pp_magic, pp_magic); -NET_IOV_ASSERT_OFFSET(pp, pp); +NET_IOV_ASSERT_OFFSET(pp_item, pp); NET_IOV_ASSERT_OFFSET(dma_addr, dma_addr); NET_IOV_ASSERT_OFFSET(pp_ref_count, pp_ref_count); #undef NET_IOV_ASSERT_OFFSET diff --git a/include/net/page_pool/helpers.h b/include/net/page_pool/helpers.h index 793e6fd78bc5..ed068b3cee3b 100644 --- a/include/net/page_pool/helpers.h +++ b/include/net/page_pool/helpers.h @@ -83,6 +83,17 @@ static inline u64 *page_pool_ethtool_stats_get(u64 *data, const void *stats) } #endif +static inline struct page_pool *page_pool_to_pp(struct page *page) +{ + struct page_pool_item *item = page->pp_item; + struct page_pool *pool; + + item -= item->pp_idx; + pool = (struct page_pool *)item; + + return --pool; +} + /** * page_pool_dev_alloc_pages() - allocate a page. * @pool: pool from which to allocate diff --git a/include/net/page_pool/types.h b/include/net/page_pool/types.h index c022c410abe3..a5ed747f6c38 100644 --- a/include/net/page_pool/types.h +++ b/include/net/page_pool/types.h @@ -142,6 +142,11 @@ struct page_pool_stats { }; #endif +struct page_pool_item { + struct page *pp_pg; + unsigned int pp_idx; +}; + /* The whole frag API block must stay within one cacheline. On 32-bit systems, * sizeof(long) == sizeof(int), so that the block size is ``3 * sizeof(long)``. * On 64-bit systems, the actual size is ``2 * sizeof(long) + sizeof(int)``. @@ -161,6 +166,8 @@ struct page_pool { int cpuid; u32 pages_state_hold_cnt; + unsigned int item_mask; + unsigned int item_idx; bool has_init_callback:1; /* slow::init_callback is set */ bool dma_map:1; /* Perform DMA mapping */ @@ -228,7 +235,11 @@ struct page_pool { */ refcount_t user_cnt; - u64 destroy_cnt; + /* Lock to avoid doing dma unmapping concurrently when + * destroy_cnt > 0. + */ + spinlock_t destroy_lock; + unsigned int destroy_cnt; /* Slow/Control-path information follows */ struct page_pool_params_slow slow; @@ -239,6 +250,8 @@ struct page_pool { u32 napi_id; u32 id; } user; + + struct page_pool_item items[] ____cacheline_aligned_in_smp; }; struct page *page_pool_alloc_pages(struct page_pool *pool, gfp_t gfp); diff --git a/net/core/devmem.c b/net/core/devmem.c index 11b91c12ee11..3ff2e0d0d1a1 100644 --- a/net/core/devmem.c +++ b/net/core/devmem.c @@ -380,7 +380,7 @@ bool mp_dmabuf_devmem_release_page(struct page_pool *pool, netmem_ref netmem) if (WARN_ON_ONCE(refcount != 1)) return false; - page_pool_clear_pp_info(netmem); + page_pool_clear_pp_info(pool, netmem); net_devmem_free_dmabuf(netmem_to_net_iov(netmem)); diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 18645907cec2..2578f37fc1e1 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -267,14 +267,12 @@ static int page_pool_init(struct page_pool *pool, return -ENOMEM; } + spin_lock_init(&pool->destroy_lock); atomic_set(&pool->pages_state_release_cnt, 0); /* Driver calling page_pool_create() also call page_pool_destroy() */ refcount_set(&pool->user_cnt, 1); - if (pool->dma_map) - get_device(pool->p.dev); - if (pool->slow.flags & PP_FLAG_ALLOW_UNREADABLE_NETMEM) { /* We rely on rtnl_lock()ing to make sure netdev_rx_queue * configuration doesn't change while we're initializing @@ -312,15 +310,91 @@ static void page_pool_uninit(struct page_pool *pool) { ptr_ring_cleanup(&pool->ring, NULL); - if (pool->dma_map) - put_device(pool->p.dev); - #ifdef CONFIG_PAGE_POOL_STATS if (!pool->system) free_percpu(pool->recycle_stats); #endif } +static void page_pool_item_init(struct page_pool *pool, unsigned int item_cnt) +{ + struct page_pool_item *items = pool->items; + unsigned int i; + + WARN_ON_ONCE(!is_power_of_2(item_cnt)); + + for (i = 0; i < item_cnt; i++) + items[i].pp_idx = i; + + pool->item_mask = item_cnt - 1; +} + +static void page_pool_item_uninit(struct page_pool *pool) +{ + struct page_pool_item *items = pool->items; + unsigned int mask = pool->item_mask; + unsigned int i; + + if (!pool->dma_map) + return; + + spin_lock_bh(&pool->destroy_lock); + + for (i = 0; i <= mask; i++) { + struct page *page = READ_ONCE(items[i].pp_pg); + + if (!page) + continue; + + WARN_ONCE(1, "dma unmapping in %s: %p for %p\n", __func__, page, + pool); + + dma_unmap_page_attrs(pool->p.dev, page_pool_get_dma_addr(page), + PAGE_SIZE << pool->p.order, + pool->p.dma_dir, DMA_ATTR_SKIP_CPU_SYNC | + DMA_ATTR_WEAK_ORDERING); + page_pool_set_dma_addr(page, 0); + } + + pool->dma_map = false; + spin_unlock_bh(&pool->destroy_lock); +} + +static bool page_pool_item_add(struct page_pool *pool, struct page *page) +{ + struct page_pool_item *items = pool->items; + unsigned int mask = pool->item_mask; + unsigned int idx = pool->item_idx; + unsigned int i; + + for (i = 0; i <= mask; i++) { + unsigned int mask_idx = idx++ & mask; + + if (!READ_ONCE(items[mask_idx].pp_pg)) { + WRITE_ONCE(items[mask_idx].pp_pg, page); + page->pp_item = &items[mask_idx]; + return true; + } + } + + pool->item_idx = idx; + WARN_ONCE(1, "pool->items array is full\n"); + return false; +} + +static void page_pool_item_del(struct page_pool *pool, struct page *page) +{ + struct page_pool_item *item = page->pp_item; + struct page_pool_item *items = pool->items; + unsigned int idx = item->pp_idx; + + WARN_ON_ONCE(items[idx].pp_pg != page); + + WRITE_ONCE(items[idx].pp_pg, NULL); +} + +#define PAGE_POOL_MIN_ITEM_CNT 512 + /** * page_pool_create_percpu() - create a page pool for a given cpu. * @params: parameters, see struct page_pool_params @@ -329,10 +403,14 @@ static void page_pool_uninit(struct page_pool *pool) struct page_pool * page_pool_create_percpu(const struct page_pool_params *params, int cpuid) { + unsigned int item_cnt = (params->pool_size ? : 1024) + + PP_ALLOC_CACHE_SIZE + PAGE_POOL_MIN_ITEM_CNT; struct page_pool *pool; int err; - pool = kzalloc_node(sizeof(*pool), GFP_KERNEL, params->nid); + item_cnt = roundup_pow_of_two(item_cnt); + pool = kvzalloc_node(struct_size(pool, items, item_cnt), GFP_KERNEL, + params->nid); if (!pool) return ERR_PTR(-ENOMEM); @@ -340,6 +418,8 @@ page_pool_create_percpu(const struct page_pool_params *params, int cpuid) if (err < 0) goto err_free; + page_pool_item_init(pool, item_cnt); + err = page_pool_list(pool); if (err) goto err_uninit; @@ -350,7 +430,7 @@ page_pool_create_percpu(const struct page_pool_params *params, int cpuid) page_pool_uninit(pool); err_free: pr_warn("%s() gave up with errno %d\n", __func__, err); - kfree(pool); + kvfree(pool); return ERR_PTR(err); } EXPORT_SYMBOL(page_pool_create_percpu); @@ -499,19 +579,24 @@ static struct page *__page_pool_alloc_page_order(struct page_pool *pool, if (unlikely(!page)) return NULL; - if (pool->dma_map && unlikely(!page_pool_dma_map(pool, page_to_netmem(page)))) { - put_page(page); - return NULL; - } + if (unlikely(!page_pool_set_pp_info(pool, page_to_netmem(page)))) + goto err_alloc; + + if (pool->dma_map && unlikely(!page_pool_dma_map(pool, page_to_netmem(page)))) + goto err_set_info; alloc_stat_inc(pool, slow_high_order); - page_pool_set_pp_info(pool, page_to_netmem(page)); /* Track how many pages are held 'in-flight' */ pool->pages_state_hold_cnt++; trace_page_pool_state_hold(pool, page_to_netmem(page), pool->pages_state_hold_cnt); return page; +err_set_info: + page_pool_clear_pp_info(pool, page_to_netmem(page)); +err_alloc: + put_page(page); + return NULL; } /* slow path */ @@ -546,12 +631,18 @@ static noinline netmem_ref __page_pool_alloc_pages_slow(struct page_pool *pool, */ for (i = 0; i < nr_pages; i++) { netmem = pool->alloc.cache[i]; + + if (unlikely(!page_pool_set_pp_info(pool, netmem))) { + put_page(netmem_to_page(netmem)); + continue; + } + if (dma_map && unlikely(!page_pool_dma_map(pool, netmem))) { + page_pool_clear_pp_info(pool, netmem); put_page(netmem_to_page(netmem)); continue; } - page_pool_set_pp_info(pool, netmem); pool->alloc.cache[pool->alloc.count++] = netmem; /* Track how many pages are held 'in-flight' */ pool->pages_state_hold_cnt++; @@ -623,9 +714,14 @@ s32 page_pool_inflight(const struct page_pool *pool, bool strict) return inflight; } -void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem) +bool page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem) { - netmem_set_pp(netmem, pool); + if (unlikely(!page_pool_item_add(pool, netmem_to_page(netmem)))) + return false; + + WARN_ON_ONCE(page_pool_to_pp(netmem_to_page(netmem)) != pool); + + //netmem_set_pp(netmem, pool); netmem_or_pp_magic(netmem, PP_SIGNATURE); /* Ensuring all pages have been split into one fragment initially: @@ -637,10 +733,13 @@ void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem) page_pool_fragment_netmem(netmem, 1); if (pool->has_init_callback) pool->slow.init_callback(netmem, pool->slow.init_arg); + + return true; } -void page_pool_clear_pp_info(netmem_ref netmem) +void page_pool_clear_pp_info(struct page_pool *pool, netmem_ref netmem) { + page_pool_item_del(pool, netmem_to_page(netmem)); netmem_clear_pp_magic(netmem); netmem_set_pp(netmem, NULL); } @@ -672,9 +771,13 @@ static __always_inline void __page_pool_release_page_dma(struct page_pool *pool, */ void page_pool_return_page(struct page_pool *pool, netmem_ref netmem) { + unsigned int destroy_cnt = READ_ONCE(pool->destroy_cnt); int count; bool put; + if (unlikely(destroy_cnt)) + spin_lock_bh(&pool->destroy_lock); + put = true; if (static_branch_unlikely(&page_pool_mem_providers) && pool->mp_priv) put = mp_dmabuf_devmem_release_page(pool, netmem); @@ -688,9 +791,13 @@ void page_pool_return_page(struct page_pool *pool, netmem_ref netmem) trace_page_pool_state_release(pool, netmem, count); if (put) { - page_pool_clear_pp_info(netmem); + page_pool_clear_pp_info(pool, netmem); put_page(netmem_to_page(netmem)); } + + if (unlikely(destroy_cnt)) + spin_unlock_bh(&pool->destroy_lock); + /* An optimization would be to call __free_pages(page, pool->p.order) * knowing page is not part of page-cache (thus avoiding a * __page_cache_release() call). @@ -1026,14 +1133,14 @@ static void __page_pool_destroy(struct page_pool *pool) static_branch_dec(&page_pool_mem_providers); } - kfree(pool); + kvfree(pool); } static void page_pool_empty_alloc_cache_once(struct page_pool *pool) { netmem_ref netmem; - if (pool->destroy_cnt) + if (pool->destroy_cnt > 1) return; /* Empty alloc cache, assume caller made sure this is @@ -1049,7 +1156,7 @@ static void page_pool_empty_alloc_cache_once(struct page_pool *pool) static void page_pool_scrub(struct page_pool *pool) { page_pool_empty_alloc_cache_once(pool); - pool->destroy_cnt++; + WRITE_ONCE(pool->destroy_cnt, pool->destroy_cnt + 1); /* No more consumers should exist, but producers could still * be in-flight. @@ -1131,10 +1238,14 @@ void page_pool_destroy(struct page_pool *pool) if (!page_pool_put(pool)) return; + /* disable dma_sync_for_device */ + pool->dma_sync = false; + page_pool_disable_direct_recycling(pool); + WRITE_ONCE(pool->destroy_cnt, 1); - /* Wait for the freeing side see the disabling direct recycling setting - * to avoid the concurrent access to the pool->alloc cache. + /* Wait for the freeing side to see the new pool->dma_sync, + * disable_direct and pool->destroy_cnt in page_pool_put_page. */ synchronize_rcu(); @@ -1143,6 +1254,8 @@ void page_pool_destroy(struct page_pool *pool) if (!page_pool_release(pool)) return; + page_pool_item_uninit(pool); + page_pool_detached(pool); pool->defer_start = jiffies; pool->defer_warn = jiffies + DEFER_WARN_INTERVAL; diff --git a/net/core/page_pool_priv.h b/net/core/page_pool_priv.h index 57439787b9c2..2f83caea7be7 100644 --- a/net/core/page_pool_priv.h +++ b/net/core/page_pool_priv.h @@ -36,16 +36,17 @@ static inline bool page_pool_set_dma_addr(struct page *page, dma_addr_t addr) } #if defined(CONFIG_PAGE_POOL) -void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem); -void page_pool_clear_pp_info(netmem_ref netmem); +bool page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem); +void page_pool_clear_pp_info(struct page_pool *pool, netmem_ref netmem); int page_pool_check_memory_provider(struct net_device *dev, struct netdev_rx_queue *rxq); #else -static inline void page_pool_set_pp_info(struct page_pool *pool, +static inline bool page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem) { } -static inline void page_pool_clear_pp_info(netmem_ref netmem) +static inline void page_pool_clear_pp_info(struct page_pool *pool, + netmem_ref netmem) { } static inline int page_pool_check_memory_provider(struct net_device *dev, diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 74149dc4ee31..d4295353ca6e 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1033,7 +1033,8 @@ bool napi_pp_put_page(netmem_ref netmem) if (unlikely(!is_pp_netmem(netmem))) return false; - page_pool_put_full_netmem(netmem_get_pp(netmem), netmem, false); + page_pool_put_full_netmem(page_pool_to_pp(netmem_to_page(netmem)), + netmem, false); return true; } diff --git a/net/core/xdp.c b/net/core/xdp.c index bcc5551c6424..e8582036b411 100644 --- a/net/core/xdp.c +++ b/net/core/xdp.c @@ -384,7 +384,8 @@ void __xdp_return(void *data, struct xdp_mem_info *mem, bool napi_direct, /* No need to check ((page->pp_magic & ~0x3UL) == PP_SIGNATURE) * as mem->type knows this a page_pool page */ - page_pool_put_full_page(page->pp, page, napi_direct); + page_pool_put_full_page(page_pool_to_pp(page), page, + napi_direct); break; case MEM_TYPE_PAGE_SHARED: page_frag_free(data);